AN INVESTIGATION OP THE RELATION BETWEEN WORK ON ASSIGNMENTS
AND ACHIEVEMENT IN A GENERAL EDUCATION SCIENCE COURSE

By
Edward Vernon Perkins

A THESIS

Submitted to the School of Graduate Studies of Michigan
State College of Agricvilture and Applied Science
in partial fulfillment of the requirements
for the degree of
DOCTOR OP EDUCATION

Department of Higher Education
1953

AM IMTBSTIOATIOH 09 THX KELATIOM BCTBJB1 S O W OB A8S KWMWT8
AMD AORIMTBCOR II A OBTBUL MDUCATIOM SOI KICK OOUR8B

H w u r d T«v m b Poaflclns

AM ABSTRACT
Suhol tt*A to tho Sohool ®f Onutaftt® SIv AIm

of Ulehign

Stitt* Oollogo of A g r l m l t v n oad Appl&oA Solooo*
la port 1*1 folfillaoat of tho roqulr— onto
for tho Aonroo of
dootob

car xo dg a t x o m

Doportaoot of R&^har IdaoatlOB

1953

Approrod__

1

Edward Tamoa P a A l a i

ah xftestigatxof of the relatiof n r a n won of ASsifflncBrrs
AFD ACHI1YBIWT IF A OBTKRAL EDQCATXOF BCIFFOF COURSE
ABSTRACT

Tho p n U «

of this & m a U f a U « B lo tho quootloas

What &■ tho

relation hotwoon eooplot&om of hoooooric aoalgHOita aid autaoquait
aohloroawat la a oollogo ooianeo eoursoT
aros

Tho attoedaat prohloaas ota&lod

(1) an examination of* tho purported fuotlosa of hoooauat asals^

aoati; (?) tho douolopaont of a ourwoy toohalqoo to ohtala aoourato data
froa ■indent* oonoornlic hoaaooaovk porforaaaoej (3) tho aaalyolo of tho

aurrojr data, hy appropriate statlatieal aothodo, to dotoraalao tho oaetoat
of tho aeewaed oorrelatlwo rolatloa between work dono oa aoolcaoowto aad
aohlaraaiant; and, (U) tho earoful appraloal of tho Taluo of hoaaoooitc
aaalffaaita la a typical olaoorooaa altoatioa where aeholaatlo growth lo
known to oocur.
Tho atadjr oonoleto of tho our way of houaoork porfonoieo of U73
oolloso seloaoo oouroo otudoato oho leapt a atudy log of asalgaaeate,
woit doao. and work roportod done during a full tana of olaoaoa.

Weakly

queetlonnalree war* used to ollolt tho data ao to porfermoaeo of tho aa—
•lgnaents.

Data froa tho q'ueetlonnalree aad data oonoornlng payehologloi

aad reading toot deellee, quartllo n a k la high oohool graduation elaso,
aad also of high oohool froa adxlah grmduatod. were tranoforrod to

•

H
9
«♦
►*
o

0

5£

N

•

O

<♦

•

r r
«► i»

9
9

*

P

pr
►*

«
►*

*

S'

JH ►I*

•

I
^

»*

M
•

O

d

9
H
d

8

0
'

8M

•
<

d

3&

H

t

.

£

it

11

9
m

o
o

►*

►*
o

Pi

t>

g

t
f
c

9

»

8

a

9

£

8

'

8«►

*

«♦

,

?

9

►%

8

H
9

8 a
3

I

•

i

►%
O

+
ro
3 w:

9
H

(k

O*
9
*

1
P
O
0

£ ! I
9
*

S

6

9

S'

^

t
9

f
•» Gt
v/ P

*
C

r
6

8

ft

*

l

r

w

2

i

8

f

£
0

\
^

I

£

Kt

9
*4

®

8

4

9

v
9

*
8

9

8

9

9

8
8

M

?

0

*
8

3

8

9
*
►»

9

0

3
9

H

O
9

&

•f
9

9
H
?
9

9

>4

8

i

9

?

•»
►*

9

!
•
8

8 3
«■ |

9

9

►4

I

•

3

3

O*
9

8
9

•0

I

tr

9
H
H

M

M

8

89

9

8*

8

t*

8

#

8

8

8

9

.

;

.

.

4

j

H
9

8

9

*

i

4

•

8
8

8

K
H*

t*

8

S •

9

8

M S I *
;
| i 3h i
h
U
n J
8 ?
I i
: i 9H
89
8
«»
9

H

M
»»

i
8
S

t*

*

9

:

it
0*
9

3
r

9

n

9

2

►4

0

*
8
9

*

«►

P

•

sr
►*

£
9

P

%

g

H

8

9 i
8
P

8
K

9
1

H

;

9

4

►%

5

9

f

r I j
1 * i

9

1
i

&

ft

1

*
1
«*

9

cr

&

ST

•
)

8

O
H

Ol

3

i
t

«
V

ft

8

t
0t
*

*
©

«*
►*
9
0

;

«*

9
1
9

d

•
«

«♦

8

I

9

m

0*

'

H

■

X9A#x 9*»xxw>

1

3
a | i
I
I?
v
f i § j 3* I
II
i
!
I
I£
rl » 8 * I ; i g i8 i
5
iu iI
\I
I
t
i
4
|
8 3
i
!
I
I
I
4
I
8
H
3
i
i
8
H
I
! hi ?
s
I
I
I
>
i
i
i I
H£i I r I 3 (I
S
£
?
I !
8
i
•

#q% pn

0

I > ‘
6
r
* 9

? I

0

•4

3
■Award Tr nra Ptikist

Aaplo

of adholaotlo provth la tho aubjoot aattor fiold of

tho iaTootl«atioa wao f o m d hat apparoatly tho woik dona oa tho uaift*
■onto lo net alooaly rolatod to tho aubaoqaaat aahlawaaaat la tho couroo.
fho loarnla* whioh tokoo plaeo appoaro to oooar la tho elaaarooa ltoolf
aad la apparoatly aoro rolotod to pogrhholocleal faotoro aad roartlng abil­
ity than to tho weiat of work doao oa out-of-olaao aoalcaaoata.

fhlo

laroaticatloa did aot ohew why tho ooascaaly aaaoaod oorrolatiTo rolatioa
hotwooa hoaowoak porforaaaeo aad aehloroaoat lo aot oorrohoratod.

ACKNOWLEDGMENTS

The writer wishes to express his thanks to the chairman of his
Guidance Committee, Milosh Muntyan, Professor and Head of the Department
of Higher Education, and to the other members, Cecil Millard, Acting Dean
of the School of Education and Graduate Council representative, Clyde
Campbell, Professor and Head of the Department of Educational Administra­
tion and Supervision, Walter Johnson, Associate Professor of Guidance
and Counselor Training, and Chester Lawson, Professor and Head of the De­
partment of Natural Science, Special indebtedness to Professor Lawson
for kind guidance and valuable help is gratefully acknowledged.
The cooperation of Professors Miriam S. Lucas and Marvin D. Solomon,
and of James M, Elliott, Instructor, all of the Department of Natural
Science, is appreciated. The investigator extends his thanks to the fol­
lowing members of Michigan State College who also rendered assistance:
John N. Moore, Department of Natural Science, Mr. Leisenrlng, Mr. Stout—
enberg, and Mr. Becfloy of Records, Registration and Admissions offices,
Nr. Martin and Miss Tasehner of tho Tabulating office.
The author wishes to express his sincere gratitude to Dr. George
Hilliard, Professor of Education, Western Michigan College of Education,
under whose inspiration and continual encouragement, the author continued
his advanced education, and to whom the study is dedicated.
The financial support of the Michigan State College All College Re­
search Fund, which made the completion of the study possible, is deeply
appreciated.

Edward Vernon Perkins
candidate for the degree of
Doctor of Education
VITA.

Final Examination:
Thesis:

March 9» 1953* 8-10 a.m., 202A Morrill Hall.

An Investigation of the Relation between Work on Assignments
and Achievement In a General Education Science Course.

Outline of Studies:
liejor subject— E i ^ e r Education
Cognate field— —Biclcgicel Science
Biographical Items:
Bora:
Decanter 16, 1917Undergraduate Studies: Albion College, 1936— 1939? Western
Michigan College of Education, 1939-19^1Graduate Studies: University of Michigan, 19^+6— 19^7» Master
of Arts in Education, University of Michigan, 19^75
Michigan State College, 19^8— 1953Experience: Teaching principal, Pinnebog High School, Pinnebog
Michigan, 19^1? Classification Interviewer and Classifica­
tion Section Chief, Personnel Technician, Chemical Warfare
Services Headquarters, 19^2-^35 Control Tower Operator,
U. S. Army Air Forces, 19^3— 19^55 Instructor, Michigan
State College department of Biological Science, I9U7— ,
Member of: American Association for the Advancement of Science
Michigan Academy of Science Arts and Letters; American
Association of University Professors.

table of

contents

ACKNOWLEDGMENTS.......................................................

I1

V I T A ................................................................... 111
CHAPTER I.

INTRODUCTION

A. The Problem
Statement of theproblem. .................................
Importance of the problem.................................

1
2

B. Definition of Terms
Homework assignment......................................
Homework performance.....................................
Achievement...............................................

7
8
8

C.

Plan of the Study..........................................

9

D.

Limitations of the Study..................................

9

S.

Organization of the Remainder of the Dissertation........

10

CHAPTER II.

REVIEW OF THE LITERATURE

A.

Purported Functions of Homework Assignments..............

12

B.

The questionnaire as a Survey Method in Education........

13

C.

The Study Log and Diary as a Survey Method in Education...

D.

Related Studies at the College Level......................

19

E.

Summary.....................................................

23

CHAPTER III.

18

THE METHODS OF INVESTIGATION OF THE PROBLEM

A. The Population..............................................
B. Official Sources of Data
Records, Registration and Admissions offices...........
Biological Science departmental records.................
Board of Examiner a............ .

lv

25
26
27

C. The Pilot Studies and the Development of the Log*Questionnaire Survey Method
Pilot questionnaire and method.......................
Pilot questionnaire findings..........................
Trial revisions of the pilot questionnaire method...*
D. The Construction of the Study Log and Questionnaire
Construction of the Study Log.........................
Construction of the Questionnaire....................
E. The Administration of the Stud;.' Log and Questionnaire to
the Population.
Administration of the Study Log......................
Administration of the Questionnaire..
..........
F. Methods of Analysis of the Data
Me clianioo.1 tabula tio n .......
.....................
Clerical and tabulation procedures...................
Statistical computation...............................
G. Some Limitations of the Method of Investigation
Limitations of the questionnaire data............... .
Minimizing errors attributable to memory.............
Correlation coefficients from re— take data...........
Interpretations from simple correlations.............
IV. PRESENTATION AND ANALYSIS OF THE DATA
A. Data and Statistical Inferences..........................
B. Summary of Statistical Inferences.......................
C« Homework Performance of the Mean Student of the
Investigation.............................................
D.

Scholastic Growth of the Population.....................

V.

SUMMARY OF FINDINGS, EDUCATIONAL IMPLICATIONS, AND LIMITATIONS
OF THE FINDINGS, WITH SUGGESTIONS FOR FURTHER STUDY
A. Scumnary of Findings
Homework assignments in the investigation..........
Homework performance in the investigation........
Relation of the amount of homework reported done
and achievement.................
Analysis of achievement................................
Counseling of students about homework.................
Scholastic growth and homework performance............

86
8J

B. Educational Implications of the Investigation.............

91

C. Limitations of the Findings with Suggestions for
Further Study...............................................

93

APPENDIX

I.

TABLE......................

APPENDIX

II. LOG— QUESTIGNUAIRE MATERIALS.........................

APPENDIX III. CLERICAL AND

TABULATION MATERIALS....................

BIBLIOGRAPHY...........................................................

vi

88
39
90
91

10?
108
113
117

LIST OF TABLES
Page
Table I.
II.

III.

IT.

T.

VI.

VII.

VIII.

IX.

X.

Pilot Study Data from 226 Students Reporting on
the Questionnaires*.................................

30

Data for Calculation and the Calculated Simple
Correlation of the Questionnaire Items from Original
and Written Re—Take.................................

*+8

Data for Calculation and the Calculated Simple Cor­
relations of the Questionnaire Items from Original
and Oral Re— Take....................................

^9

Data for Calculation of Simple Correlations of Home­
work Performance with Term Grades

5^

Data for Calculation of Simple Correlations Between
Tern Grades* Certain Indices of General College Abil­
ity (Cooperative Test of Reading Comprehension Decile
Rank, American Council Psychological Examination Q*
L, and Total Score Decile Ranks). and Class of High
School from which Graduated......
55
Data for Calculation of Simple Correlations of Home­
work Performance with Comprehensive Examination
Grades................................................

6l

Data for Calculation of Simple
Correlations Between
Comprehensive Examination Grades and Cooperative
Test of Reading Comprehension Decile Rank, American
Gounci1 Psychological Examination Q, L, and Total
Score Decile Rank, and Class of High School from
which Graduated......................................

62

Data for Calculation of Simple
Correlations of
Homework Performance with Decile Rank as Determined
by the Cooperative Test of Reading Comprehension
Scores................................................

6U

Data for Calculation of Simple
Correlations of
Homework Performance with Decile Rank as Determined
by American Council Psychological. Examination
Q Score...............................................

65

Data for Calculation of Simple
Correlations of
Homework Performance with Decile Rank as Determined
by the American Council Psychological Examination
L Score...............................................

66

vii

Table XI.

Data for Calculation of Simple Correlatione of
Homework Performance and the A^.oricrn Council
Psychological Examination Total Score Decile
Bazik....................................................

67

Data for Calculation of Simple Correlations of
Homework Performance with Quartile Rank in High
School Graduation Class...............................

68

Data for Calculation of Simple Correlation of
Homework Performance with Class of High School from
which Graduated........................................

69

Summary of Correlation Coefficients Computed for
the Analysis of Achievement...........................

73

Summary of Correlation Coefficients Computed for
the Analysis of Homework Performance.................

7^-

XVI.

Intercorrelations of Homework Performance.............

"J8

XVII.

Grade Equivalents of Scores Made by Students on a
Biological Science Pre-Test and on Comprehensive
Examination............................................

82

XII.

XIII.

XIV.
XV.

XVIII. Measures of Central Tendency and Dispersion of
Pre-Test Scores of Entering Freshmen and Scores
of a group of Students on the Same Test after Com­
pleting the Course.............
XIX.

85

Distribution of Term Grades and Homework Performance
for the 28 Respondents in the American Council Psy­
chological Examination Tenth Decile Group.......

97

Distribution of Term Grades and Homework Performance
for the 30 Respondents in the American Council Psy­
chological Examination First Decile Group............

98

Distribution of Term Grades and American Council
Psychological Examination Deciles of a Random
Sample of 178 of the Respondents According to the
Number of Hours Per Week of Study....................

99

FIGURE I. The Mean Student of the Investigation....................

SO

XX.

XXI.

vi 11.

1

CHAPTER I
INTRODUCTION
A. THE PROBLEM
Statement of the problem.
question:

The problem of this investigation is the

What is the relation between completion of reading and problem

assignments which instructors give to students to be done outside of
class and trie subsequent achievement in a college science course?
Teachers, students, administrators, and parents have accepted the
-orevalence of the assignment of work to be done outside of class, in
American schools and colleges.

Teachers and institutions which have not

incorporated the assignment of outside reading and problems as a regular
teaching method have been exceptions to such an extent that anonymity or
notoriety have accompanied such a deviation from the usual procedure.
Teachers have employed this device as if the positive correlative rela­
tionship between performance of such outside work and subsequent achieve­
ment in the course were a foregone conclusion.
ing method, based on such an assumption,

A widely oracticed teach­

should be subjected to investi­

gation to discover the values and limitations of the method in education­
al institutions.
The problem has been delimited in this study to the investigation of
the relation between the amount of the assigned outside work the students
do for a given college science course and their subsequent achievement in
that course.

The design of the study includes the following:

(1) the

examination of the ouruorted functions of outside assignments in educational

2

methodology;

(2) the survey of the literature related to the problem;

(3) the development of a survey technique and its use with students in
a college classroom situation to obtain data as to the amount of work
done on outside assignments;

(U) the analysis of the survey data, by

appropriate statistical methods, to determine the extent of the quan­
titative relation between performance on the outside assignments and
subsequent achievement in the course as indicated by grades or marks;
and, (5) the limitations of the investigation, along with recommendations
for further study of the problem.
Importance of the quantitative Investigation of the homework as­
signment performance in relation to a chievement.

Homework has been

assigned by so many teachers in so many teaching situations for so many
years that no one appears to question very seriously the contribution
of homework to achievement.

Teachers in the typical classroom situation

apparently make homework assignments with the expectation that the stu­
dents will carry them out.

One assumption, inherent but seldom uttered,

is that the amount of performance of the homework makes a positive and
significant contribution to the subsequent achievement of the students.
The assignment of homework seems to be accepted as one of the methods
for bringing about learning on the part of students.
Published investigations have disclosed no evidence to indicate the
extent to which students do the assigned homework.

The assignment of

homework has not been questioned seriously by many authors of education­
al methods textbooks.
n

These authors have customarily devoted space to

discussion of assignments and have given more than casual emphasis to

3

it.

Even those teachers and authors who stressed the importance of

homework for student learning have failed to produce evidence which
substantiates the quantitative assumptions underlying the assignments*
Woodring, in a general discussion of assignments,

says:

Of all the resoonsibilities which the teacher must face, the
assignment is the most important.
It is the pivot of suc­
cessful teaching.... With the assignment she sets the stage
for action and controls the amount and kind of turpil activity.
It is the crux of the study problem* The amount of time
spent in study and the quality of the preparation will be
conditioned by the kind of assignment given.
Woodring'8 statement in support of the assignment as an important activ­
ity of the teacher does not include any substantiation of the claim.
Such statements presume that the pupil activity produced as a result of
the assignment makes a significant contribution to student learning*
A similar discussion of assignments by Yoakum describes certain
attributes which would seem to make the assignment indispensible to any
learning situation.

It states, In part:

The assignment is still fundamental.— Assignment of lessons
is still fundamental in teaching.
The practice of assign­
ment, at least as to theory,,is changing very rapidly, but
fundamental princloles of value still remain. The assign­
ment is important because it is the vital activity in the
pursuit of learning, it provides an opportunity for creating
mental set, it includes the aopraisal of knowledge, it makes
necessary knowledge concerning organized procedures in teach­
ing and learning, it requires a consideration of individual
differences, it necessitates plans for socialized living and
learning.«

H j. N. Woodring and C. W. Fleming, "Directing Study Through the
Assignment," Teachers College Record, 33*673» May, 193?»
A. Yoakum, The Improvement of the Assignment (New York:
The
MAcmillan Company, 193<?)» P»

While Yoakum gives recognition to desirable characteristics of assign­
ments, a lengthy discussion is included without reference to any study
which reports evidence in support of the stated contentions.
Some arguments offered for the assignment method are:

to assure

that students will go in the desired direction, and to make a course
more challenging.

Such arguments, bearing the mark of a typical homework

assignment, are suggested by Luella Cole:
Assignments are the student's sailing directions.
If they are
confused or boresome, he either will not know where he is
supnosed to go or will not care whether or not he ever gets
there. Althou^i not every assignment needs to be o r i g i n a l after all a teacher is only human and cannot be expected to
have inspirations daily— a dash of ingenuity and novelty in
assignments adds greatly to the value.
It is the consensus
of opinion that the progress of students in a course bears
direct relation to the degree of challenge in the work re­
quired of them.-5
The discussions of assignments offered by Woodring and Fleming,
Yoakum, and Cole omit one very important consideration.

For, even if

the assumption were that assignments could possibly accomplish the out­
comes stated, evidence is still lacking in the discussions to indicate
that the amount of the assignments performed makes a significant contri­
bution to achievement in the course.

An example of the reasoning of an

authority on college teaching is that of Cole in the statement, "The
value that students derive from a course is usually in direct ratio to
the kind of work they are asked to do.

On the college level there has

been almost no research in this matter, probably because supervisors

■^Luella Cole, The Background for College Teaching (New York:
and Rinehart, Inc., 19**o), p.

Farrar

5
h.

do not walk into college classes fluid listen...."

The lack of research,

concerning assignments and outcomes Indicates one reason for the study
of a teaching method which is so nearly universal in American schools
and colleges.
The writer*a survey of fifty educational methods textbooks Indicates
that teacher training institutions apparently continue to include the as­
signment of homework as a suggested teaching method.

The analysis of the

fifty educational methods textbooks listed in the Table in Appendix I, re­
veals that while 21, U33 pages in these texts are devoted to the various
aspects of teaching, only 1,079» or five per cent of the pages were listed
in the indexes under "assignments."

Apparently the authors of these

books felt that assignments were essential enough to Include some men­
tion of them in a methods textbook, but, with the exception of the two
who wrote complete volumes on the subject of assignments, these authors
devoted only a very small percentage (five per cent) of the pages to the
enlightenment of students of teaching methods concerning assignments.
Most of the discussions of assignments in these texts deal with the "how"
of assignments but do not mention "shy" assignments are essential, "why"
one assignment is more effective than another, or "what" the assignment
actually accomplishes for the student.

The complete omission of sub­

stantiating evidence for homework assignments to which these authors
commit themselves is aost unfortunate because no basis is provided for
Judging the values and limitations of homework assignments and the ex­
pected outcomes.

This lack of evidence points up the importance of an in­

vestigation of homework assignment performance in relation to achievement.
TT-----

Ibid. . p. 36?.
»' V •
f,:h.

6

Any Investigation of the assignment of homework as a teaching
method should attempt to evaluate the method by examining the results
of the method as indicated by student achievement*

The literature found

relative to homework assignments* performance on homework, and achieve­
ment, reported no investigations, at the college level, of the assumed
"cause— offeet" relationships between the amount of homework performance
and the subsequent achievement by students as indicated by marks or
grades*
The need for an investigation of the relation between performance
on homework assignments and achievement is Indicated also by the counsel­
ing functions where a diagnosis of student achievement in relation to
performance of homework is attempted frequently.

The counseling of stu­

dents concerning modification of performance of homework assignments in
anticipation of improved scholastic performance is frequently attempted
by teachers and counselors who assume laggardlinesa to be the cause of
deficient scholarship.

The assumption is expressed in the advice that

"There*s no reason why you can't do better if you*11 Just work harder
and longer."

While laggardliness may be one of the causes of poor a-

chievement in individual cases, there are no studies which show that it
18 the moat frequent or most consistent cause of poor achievement in
schools and colleges.

An investigation of the relation between per­

formance of homework assignments and achievement is necessary before the
counseling of students for the modification of homework performance can
become consistently effective and reliable.

t'

7

Improvement of instructional methods may be aided by a careful
evaluation of relations between achievement and various factors of
classroom activities.

These might include the students' reading and

psychological aptitudes, quartile rank in high school graduation class,
academic growth in subject-matter courses, and the size of high school
from which the strident graduated.
The importance of an investigation of homework performance in rela­
tion to achievement in American schools and colleges is derived froa (1)
the lack of evidence to corroborate the assumed high and positive rela­
tionship between the amount of performance on homework and subsequent a—
chievement as indicated by grades or marks;

(?) the prevalent practice

of the assignment of homework, as usually propounded by some educational
textbook authors with no pretense of substantiation;

(3) the inadequacy

of evidence upon which to base counsel of students to modify their study
hAbits in anticipation of more performance of homework and consequent
improved scholastic achievement; and, (U) the continuous study of educa­
tional practices to aid the improvement of instruction.
B.
Homework assignment.

DEFINITIONS OF TERMS
The word "assignment" has been used to de­

scribe almost everything that students do or are expected to do to en5
hance learning. The meaning of homework assignment in this study is

5
A description of assignments may be found in W. 3. Monroe, ed.,
Encyclopedia of Educational Research (New York:
The Macmillan Companv,
1950), p. 7 % .

4

8

limited to any and all matters within the course to which the Instructor
dictates the students shall direct their attention outside of class.
Homework performance.

Performance of homework assignments in this

study has been defined as including only certain activities outside of
class which the participating students actually reported on weekly ques­
tionnaires in class.

The reports of reading assignments completed and

the number of assigned problems completed, together with the number of
hours spent doing homework, have been utilized as the indices of home­
work performance.

Achievement.

Regardless of the faults, limitations, and errors

often attributed to marks or grades as indicators of scholastic achieve­
ment, grades and marks are used commonly in the schools and colleges of
America.

The local variations of the letter systems and the associated

numerical point systems are numerous but in all cases the marks and
grades ar? used for extra-curricular activity qualification, scholastic
recognition, and promotion.

The scholastic grade or mark, together with

the homework assignment, are the trademarks of the classroom methods of
today’s schools and colleges.

Inherent in such methods is the assumption

that the grade or mark given to a student is a reasonable indication of
his achievement.

9

C. THE PLAN 07 THE STUDY

The plan of the study Involved* essentially, a survey method of re­
search.

Students in four lecture sections of a Biological Science course

were surveyed by a questionnaire method to obtain data as to the percent­
age of the out—of— class reading and problem assignments they completed.
Psychological and reading test score deciles, quartile rank in high school
graduation class, and the class of high school from which graduated* were
recorded for each questionnaire respondent.

Simple correlations wore

computed between the variables of homework performance reported, achieve­
ment, and various indices of general college ability.

Statistical infer­

ences were drawn from the correlation data computed for the analysis of
achievement in this study.
The findings of the study permitted a partial answer to the question
being investigated, within the limitations of the survey method used to
obtain the data.

D. LIMITATIONS OF THE STUDY
The findings of this study were based upon, and stated within the
limitations of, the assumed validity of the data obtained from the
questionnaire method.
The quantitative nature of the questionnaire data used in this
study did not permit analysis of the quality of the homework performance.

10

The statistical inferences drawn from the study did not permit conclu­
sions as to why the relationships between the variables studied did or
did not appear.

Heither did the data indicate which of the variables

studied contributed most to the students* achievement in the course.

S. ORGANIZATION OF THE REMAINDER OF THE DISSERTATION
Chapter II is the presentation of the review of the literature
related to the historical and current status of the investigation of
the assumed relation between the amount of homework done and scholastic
achievement.
In Chapter III the methods of investigating the problem are de­
scribed in detail including (A) the population;
data;

(B) official sources of

(C) the pilot studies and the development of the log-questionnaire

survey method;

(D) the construction of the study log and questionnaire

us 1 in the study;

(E) the administration of the study log and question­

naire to the population;

(F) methods of analysis of the data; and, (G)

some limitations of the method of investigation.

11

The presentation and analysis of* the data are reported In Chapter
IV In four parts:

(A) data and statistical Inferences; (B) summary of

statistical inferences;

(C) the mean student In the investigation; and,

(D) the scholastic growth of the population.
The summary of findings, educational implications from the study,
and the limitations of the findings with suggestions for further study,
are to he found in Chapter V.

12

CHAPTER II
REVIEW OP THE LITERATURE
A. PURPORTED FUNCTIONS OF HOMEWORK ASSIGNMENTS
One generally accepted definition of homework assignment "la sim—
1
ply the laying out of a task and procedure for mental work,"
The as­
signment of tasks to he done outside of the classroom Is a prevalent
technique of teaching at the college level.

Noting the dearth of ob­

jective data to support what has been termed a unanimous belief of ed­
ucational methods authors in the value of homework assignments* Bossing
submitted what was purported to be corroborative evidence of the essen­
tial significance of homework assignments.

What was termed an "unbroken

emphasis" upon the assignment of homework consisted of only five refer­
ences in support of the contention of "essential significance" which
were published over a period of twenty— one years.

Bossing offered no

explanation as to why homework assignments had not been subjected to
more careful scrutiny.
The published studies of the assignment of homework which were
read have revealed no experimental evidence in support of this teach­
ing method in relation to academic achievement.

Aside from the contra­

dictory findings relative to the amount of study time spent by college
students and the resulting achievement no evidence was found from which
a positive

correlative relationship between the amount of performance of

^Nelson L. Bossing, Progressive Methods of Teaching in Secondary
Schools (Boston: Houghton Mifflin Company, 19^?!), pp. 270-271.

13

homework assignments and achievement in college could "be determined or
even inferred*
B.

THE QUESTIONNAIHE AS A SURVEY METHOD IN EDUCATION

Considerable precedent exists for the use of carefully designed
and administered questionnaires as an historical method of research.
Davis and Barrow, after analyzing the application of questionnaires to
educational research, concluded that "The questionnaire will probably
always be one of the chief means of gathering data on certain problems
in education in spite of the fact that its critics increase year by
year.1*
Support of the questionnaire as a research tool is found in Kelley*s
writings about questionnaires in the function of the scientific method
in research and education:
Instead of a single scientific method there are no less
than four methods of research*
The most pervasive is that
of dialectic or logic.
It finds its finest expression in
human life in the wonderful pronouncements of pure mathemat­
ics
Second in definiteness nd rigor is the experimental
method*, third, the histories : and last, the method of future
estimation....
If facts of a unique nature known only to specific indi­
viduals are desired, then an oral or written questionnaire
addressed to the one person knowing these facts...is surely
the only method of approach.
It is in fact an historical
method.3
*

.

■

■

^Robert Davis and Edwin L. Barrow, "A Critical Study of the Ques­
tionnaire in Education," Educational Administration and Supervision,
?l:lUU, February, 1935*
^T. L. Kelley, Scientific Method: Its Function in Research and Ed­
ucation (Columbus: ¥he Ohio State University Press, 19?9), pp. l£,

1%

The questionnaire method of gathering date, pertinent to performance
of homework assignments is the only method found in the literature.

Di­

rect observation of the study hahits of students could provide data under
some circumstances, but since techniques for such direct observation of
study habits have not been published, the questionnaire must be resorted
to for the acquisition of the datel^bout performance of homework known
only to the students.

The literature provided several criteria for the

construction and administration of questionnaires which would enable the
researcher to secure useable data.

The researcher should assure himself

that (1) he cannot collect the data except through the questionnaire, and,
U
(2 ) the problem is not trivial.
Norton coi.cl ;<;ed. after a comprehensive survey and study of the
questionnaire movement, in part, that •'...this method of investigation,
in spite of its abuse, has yielded far more on the credit than on the
5
debit side of educational advance."
A more constructive suggestion was made in a list of the criteria
of a good questionnaire:
1.

It Bhould be within the comprehension of those who are
to answer it.

2.

It should demand a minimal amount of writing.

3.

It should be directed primarily to matters of ascertain­
able fact and less often to matters of opinion.

^B. R. Buckingham, "The Questionnaire," Journal of Educational Re­
search, lU:5^— 5®, June, 1926.
5
-V. K. Norton, "The Questionnaire," Research Bulletin of the Nation­
al. Education Association. Vol. VTII, No. 1, January, 1930. p. 8.

15

U,

It should elicit unequivocal replies, especially if these
are later to he subjected to statistical treatment*

5*

It should deal with matters that are worth Investigating
and that will seem to the recipients to be worth investi­
gating.

6*

Though demanding only brief replies, it should stimulate
supplementary communications from the recipients*

7*

promise the respondent a copy of the published

The researcher will find
criticisms which are commonly

it useful to be well informed about
made of the questionnaire method

the
in order

to provide, as far as possible, the means by which such criticisms may
be minimized.

Three principal criticisms were reported by Symonds as a

result of a review of the methods that had been reported for surveying
and inventorying student study practices:
1*
2,
3*

Pupils are not trained obnervors of their own acts.
The questions are apt to be suggestive,
Pupils may definitely mislead.'

Symonds claimed that the other methods which should yield more reliable
data are:

(l) direct observation and (?) testing.

No precedent or sug—

gested method for utilizing these more desirable methods of investiga­
tion was found in the literature, however.

G. M. Whipple, MThe Improvement of Educational Research," School
and Society, ?6:?53» August, 19?7»
Fercival M. Symonds, "Methods of Investigation of Study Habits,"
School and Society, ?U:15?, July 31. 1926.
Loc, clt.

16

Another type of criticism frequently made of questionnaires was
reported b y Hubbard:
Stoke and Lehman concluded that the questionnaire is pecul­
iarly vulnerable when employed for collection of personal
information or when used with subjects who see (or imagine
they see) an opportunity to advance their personal interests
by means of the returns made by them.... Students in seven
classes in college psychology were asked to report the num­
ber of times they had taken books from the reserve desk for
assigned work.
It was possible to check the replies at the
library desk.
It was found that seven in eight students over­
estimated the number of check-outs.
Studente of A
and B
^
scholarship exaggerated the least; C and D students the most.
From this dRta Stoke and Lehman drew the conclusion that, with regard
to

questionnaire reports, one could not rely upon the statements

students as to the amount of time given to study.

of

It is not difficult

to see why the investigators might make such an inference.

But to

stote such a conclusion seems unreasonable since the study reportedly
dealt with the frequency of library book check— cuts and not the amount
of time studied.

Whether or not proper rapport had been established

during the investigation was not reported; this factor alone could account
for the aooarent bias of the students'

reports.

The investigators ac­

tually reported little basis for placing the blame for the bias upon the
questionnaire instead of upon some other variables in the study.
Apparently, the validity of the data which may be obtained by
questionnaires has been enhanced when factual and objective responses
were elicited.

Q

Monroe noted a study by Bain which "revealed that the

F.
W. Hubbard* "Questionnaires," Review of Educational Research.
9:50U, December, 1939»

17

more objective the fact inveBtigated and the more within the individual* s
10
first—hand experience the more dependable the information."
11
A eimple-recall questionnaire was used by Smith for a study in
which the memory factor could have accounted for some of the overstate­
ment which was observed.

The validity of responses involving judgmental

and opinion responses generally seemed to be less than the validity of
the factual responses.

Even the factual responses involving personal

accomplishments seemed to be somewhat colored by overstatement, however.
Another evaluation of the questionnaire method may be obtained by
observing the consistency with which the data is elicited.

Gerberich,

in relating the findings of an investigation of the consistency of
questionnaire responses, observed a consistency of JS per cent for one—
day Intervals, 7^ ter cent for ten— day intervals, and the per cent of

1?
consistency of women to be slightly higher than that of the men observed.
Similar

findings

of consistency

of responses were published

by

Cavan after submitting the questionnaires to 1?3 grade school punils at
an interval of one week in which complete ci-reement on 8J per cent of
the q\iestions was found, with the factual questions about self evoking

10W. S. Monroe, ed., Encyclopedia of Educational Research (New
York:

The Macmillan Company, 1950), p. 950.

^ ^ r a n c i B F. Smith, "The Direct Validation of Questionnaire Data,"
Educational Administration and Supervision, 21:575, November, 1935.
1?
J. E. Gerberich, "A Study of the Consistency of Informant Re­
sponses to Questions in a Questionnaire," Journal of Educational Psy­
cho logy. 3^!3°3» May, 19^7.

18

87 P© r cent agreement, and 78 oer cent agreement was found on the ques13
tions Involving attitudes toward self.
C.

THE STUDY LOG AND DIARY AS A SURVEY METHOD IN EDUCATION

The collection of data for the analysis of students* study per­
formance on homework assignments "by using a study log or diary was
found to have precedent.

Ryans computed study ratios from study logs

obtained by requiring each student in the investigation to keep for one
full week an all-inclusive record of the time devoted to studying for
college courses.

Mimeographed record blanks carrying the necessary in­

structions for their use printed on them were distributed and collected

lU
in order to facilitate the task.
In a similar study Williamson modeled a record form after the North15
western time sheet prepared by L. B. Hopkins. Freshmen in college record­
ed for one week the actual distribution of their time by using a record
form on which all of their various activities and the amount of time de­
voted to each could be written.

^ R u t h S. Cavan, HThe Questionnaire in a Sociological Research
Project,** American Journal of Sociology, 38:725* March, 1933.

lU

D.
G. Ryans, "Some Observations Concerning the Relationship of
Time Spent at Study to Scholarship and other Factors," Journal of Educa­
tional Psychology. 30:372-377. May, 1939*
15
Z. G. Williamson, "The Relationship of Number of Hours of Study
to Scholarship," Journal of Educational Psychology, 26:682-688, December,
1935.

19

Crawford collected data regarding the actual distribution of the
students* time by having Tale undergraduates record what they did for

16
one full week of school.
A variation of the study log or diary was afforded by May in edu­
cational psychology classes when students were asked to estimate the
number of hours per week that were given to study.

The reports were

made at the beginning and again at the middle of the first semester of
college.

The coefficient of reliability was then computed between the

two series of reports as 0.86.

It was possible to infer from this co­

efficient that the estimates may be reliable but no other validity es—
17
timate was reported.
D. BELATED STUDIES AT THE COLLEGE LEVEL
Williamson reported the results of a study at the University of
Minnesota as being definite enough to warrant the counseling of stu—
18
dents for certain study habit deficiencies.
Two methods of studying
the relation of the number of hours studied to resulting achievement
were reported in the above study.

In one method* the average number of

hours of study per week by the upper, middle, and lower third levels of
scholastic ability according to intelligence were computed.

A decrease

in the number of hours of study with an ihcrease in level of academic
intelligence was reported with this method by Jones and Ruch, according

A.
(New Haven:

B. Crawford, Incentives to Study: A Survey of Student Opinio
Tale University Press, 1929), p T 17.

■^M. a. May, "Predicting Academic Success," Journal of Educational
Psychology, lU:h? 9— U^O, September, 19?3*

1SW11 liam 8on, op. clt.. pp. 682— 688.

20

to Williamson.

Jones and Ruch also reported that an increase in the

number of hours of study is accompanied by a decrease in scholarship.
Williamson concluded that if we assume the study records are valid, it
seems that for students of inferior mentality to study more hours each
week does not compensate for their handicap of lower mental capacity.
Coefficients of correlations were computed in the other principal
method of determining the relationship between hours of study and a<chievement.

Three extensive studies of this type reported by Williamson

Justified the conclusion that beyond a minimum number of hours of study,
which will vary with each student, the hours a student studied has less
significance than his academic intelligence in relation to achievement.
More hours of study will be needed by students of low ability but an in­
crease in the number of hours of study will not necessarily bring about
higher scholarship.

Williamson concluded:

Experience in counseling students leads one to conclude that
a minimum of eighteen to twenty hours and a maximum of thirty
to thirty-five hours of study a week should permit students 1Q
to get the grades that their academic aptitude makes possible.
Monroe, after reviewing the work reported by Woodring and Fleming,
concluded in part, "It is interesting to note that under the prevailing
conditions of instruction not many differences appeared in the habits of
study and methods of work of bright and drill pupils."

It was also re­

ported that students who achieve better grades do not necessarily have

19Loc. cit.

21

the study habits often attributed to good students; apparently there is
no pattern or set of study habits which are invariably associated with
20
academic success or failure.
The various studies of the relation between the amount of time
« ^
spent doing homework and achievement, In institutions of higfa/veduca-

tion, have been of a very limited nature.

"Investigations of the actual

study habits show conclusively that the amount of time spent on a course
outside the classroom averages hardly more than the number of class hours
21
according to Ewer. This amount of time would be considered inadequate
for achievement to the upper limits of students' capacities as indicated
by the results of some of the investigations.
Wade, after observing the time spent by first-year college students
in study and comparing the study time with that expected by teachers in
a state normal school, reported the study time to average slightly more
for
?2
than one hour^each one-hour recitation period* Neither Ewer nor Wade
contributed any evidence to substantiate the assumption that the
amount of time used for study has a positive correlative relation with
the achievement with which the study time is usually associated.

20
Monroe, op. clt., p. 3?1«

21B. C. Ewer, College Study and College Life (Boston:
Badger, 1917)* p. 81.

Richard Q.

Op

N. A. Wade, "Comparison of the Time Spent by First-Tear Students
and Expected "by Teachers in a State Normal School," Journal of Education­
al Research, 19:183-137, March, 19?9.

22

The lengthiest series of observations of study time and scholar­
ship w e 8 reported by Ryans*

In the study of the problem, Ifcrans had

students report the amount of time spent on study for all of their col­
lege classes during one full week end the study time on a limited assign­
ment was also reported*

The reports were correlated with achievement as

indicated by the course grades of the s tudents making the reports.

Each

of the two variables correlated positively with achievement but neither
showed apparent relationship with intelligence test scores,

Ryans in­

ferred that "The present study verifies the work of at least one pre­
vious investigator in noting study time to be a significant factor in
the earning of grades.

Other attempts to clarify this problem by Craw­

ford, Jones and Ruch, sometimes have resulted in contrary findings." The
conclusions drawn by Ryans from the reported series of observations were,
in part, that:
As judged from the statistical treatment of available data
time spent at study, while, in general, unrelated to intell­
igence, is an important factor in determining the success or
failure of a student in his academic e n d e a v o r s . ^3
Unfortunately the reported attempts to analyze achievement in terms of
study time have not provided data from which the extent of the contribu­
tion of students' study time to subsequent achievement can be ascertained.
Although one may infer from such studies that the amount of time studied
is an important factor, the predictive value of the knowledge of the

amount Of study time is not known*

—

^Hyans, op. cit., p. 376.

23

X. STMMAHY
Ample precedent was discovered for using the questionnaire and
study log or diary as instruments of historical research providing that
certain criteria are met by the instruments and method of administra­
tion, and that the conclusions drawn from the questionnaire data take
into adequate consideration the krown limitations of the method.

The

literature indicated a definite need for more thorough-going means of
application of the questionnaire method in obtaining data to assure
more reliable results.

Although direct observation, experimentation, and

measurement are still preferred as methods of research, the question­
naire remains as the only instrument which will elicit certain data re­
quired for certain aspects of educational research.
Most of the studies of homework assignments have been reported
from elementary and secondary school observations.

"The gist of the

research evidence is none too favorable to assigned homework," accord—
2b
ing to Monroe.
The conclusions reported from the studies are:
(a) there is a vory small relationship between the amount of
time spent in home study and pupil progress;
(b) homework
is not significantly related to achievement as measured by
teachers’ marks or standardized tests; (c) homework at the
elementary- 8chool level has a positive relationship to suc­
cess in high school; (d) voluntary homework has about as
many values as compulsory homework; (e) the benefits of as­
signed homework are too small to counter—balance the disad­
vantages, especially for pupils in poor homes; (f) compul­
sory homework does not result in sufficiently improved aca­
demic accomplishments to Justify the retention of the "aehievement argument" as the chief Justification for homestudy assignments."^

^JJonroe, op. cit., p. 3^°
? 5lbld. , pp. 380-381

2U

Despite the generality of conclusions against the continued use of the
conventional homework assignment, many teachers still assign compulsory
homework regularly and then leave the expected performance of the assign­
ments to the whims and pursuits of the students outside of the classroom.
n The trend of the past several decades has teen the increase of time de—

26
voted to directed study in the classroom."
The published studies which provide the historical background of
the investigation of the relationship between the amount of performance
of homework assignments and subsequent achievement of college students
have been limited to observations of amount of study time and achieve­
ment.

Contradictory findings have been reported as to the kind and ex­

tent of such a relationship between the amount of study time and achieve­
ment.

The extent to which the amount of performance of homework assign­

ments contributes to achievement apparently has not been investigated at
the college level.

----

Ibid., p. 1?30.

CHAPTER III
THE METHODS OF INVESTIGATION OF THE PROBL1H
A. THE POPULATION
Respondents to the questionnaires used for the present study were
members of four Biological Science lecture sections enrolled at Michi­
gan State College, The Basic College, for the third, or spring term,
March 1951 to June 1951* the final term of the course.

Each of the four

participating lecture sections was registered with an instructor who
also met the students from his lecture section for a two—hour laboratory
class period once a week in addition to two one-hour lecture sessions
each week.

Since no factors of selection of students for the lecture

sections in the department were known to he operating, it has been as­
sumed in this 8tody that registration of the students in the four lecture
sections was at random.
Of the respondents, 280 were males and 193 were females.

Freshmen

numbered 3^5* 5^ were sophomores,IS were Juniors, two were seniors, and
the balance of lU were terminal program students in a two-year program.
B.

OFFICIAL SOURCES OF DATA

Records, Registration and Admissions offices 1 data.

Certain confi­

dential data which the Records, Registration and Admissions offices util­
ize frequently for administrative reports were available in a composite
alphabetical list of all students who had been enrolled during the current
school year which ended Just prior to the beginning of the analysis of

26

data for this study.

The amount of time which such a list saved in the

compilation of data is "beyond estimation and the extensive cooperation
of the administrative offices which compile and utilize such a list is
invaluable to the educational investigator.

The fact that the list is

compiled "by IBM (International Business Machines) machines printing from
Hollerith cards which have been verified, attests to the accuracy of such
data.
The composite list from the Records* Registration and Admissions
offices was used as the source of the data listed below:
High school quartlle rank; 1st., 2nd., 3rd . » or Uth.
Class of school from which graduated; A— 800 or more students;
B- 325— 799 students; C- 150-321*. and D- less than 150.
Bjplogical Science departmental records.

Permanent Biological

Science departmental records of a limited amount of confidential data
pertaining to students who have been enrolled in the course provided
some of the information used in this study, including:
Laboratory section number.
Term grade assigned by the instructor.
Instructors assign term grades to each student based on the stu­
d e n t s performance on tests which the instructor had constructed and
administered during the term of classes.

The tests are made up by the

instructors to be objective and to cover the assignments which have been
given to the students.

The term grades are reported by the instructor

to the departmental office where the grades are recorded in the permanent
file of class lists of students enrolled in the course.

27

Board of Examiners' records.

The Board of Examinera functions under

a head responsible to the Dean of the Basic College.

Examiners who have

the responsibility of constructing examinations for the Basic College de­
partments are selected on much the same basis as the staff hired In the
teaching departments except that some experience and real interest in ex­
amining are required.

The Board of Examiners is also responsible for the

administration of psychological, reading, and other entrance examinations
and for maintaining and distributing records of the results.
▲ comprehensive examination over the entire three terms of the Bio­
logical Science course determines the course grade for the entire year's
coursework.

The Biological Science examiner submits examination items to

a departmental committee for criticism and editing before the examination
is compiled.
Board of Examiners' records which were on file in the Biological
Science departmental office furnished the following data in this study:
Comprehensive examination grade.
American Council Psychological Exam Q, decile (Fall. 1950).
American Council Psychological Exam L decile (Fall, 1950).
American Council Psychological Exam total score decile (Fall,1950).
Cooperative Test of Reading Comprehension total score decile
(Fall, 1950).
The American Council Psychological Examination and the Cooperative Test
of Reading Comprehension are administered by the Board of Examiners to
all freshmen upon entrance to Michigan State College.
Achievement as indicated by the grades the students received from
performance on the comprehensive examination is included in this study
with the expectation that the correlations between the performances on
homework and the comprehensive examination grade would be lower than

28

those "between correlations of the performances on homework and the
grades for one term.

These lower correlations were expected "because

the comprehensive examination covers three terms of the course while the
data on the homework performance pertains only to one term of the
course.

The high coefficient of reliability reported by the Board of

Examiners for the comprehensive examination used in this study, r s 0.89*

1
is considered by the examiners to indicate satisfactory reliability.
Arguments have been made that pencil and paper tests can not and
do not measure achievement.

Nevertheless the most casual or intense

observation of the typical educational institution has forced even the
f068 of tests and measurements to concede that tests are the most common
indicators of academic achievement which exist today.
Data concerning the subject-matter knowledge of the respondents of
tliis study upon entering the first term of the course, and upon the com­
pletion of the three terms of the course, were also obtained from the
Board of Examiners.
C.

THE PILOT STUDIES AND THE DEVELOPMENT
07 THE LOG— QUESTIONNAIRE SURVEY METHOD

Pilot questionnaire and method.

Since the principal problem of

thi8 study involved performance of homework in relation to achievement,
it was necessary to obtain data about the homework performance from the

^ o a r d of Examiners, Michigan State College, Comprehensive Exami­
nations. in a Program of General Education (East Lansing:
Michigan State
4ollege Fr^ss” 1§U9), p. 1^5»

29

students.

Because of the impracticality of direct observation of stu­

dents' study performance it was desirable to utilize an instrument to
obtain the data.

Careful study of questionnaire construction and ad­

ministration provided only a meager guide for the utilization of such
an instrument for research.

A pilot questionnaire was multigraphed and

administered in an effort to gain first-hand experience with the in­
strument and method.

The pilot study data were elicited from 2?6 stu­

dents regularly enrolled at Michigan State College in the Biological
Science course of the Basic College as shown in Table I.
Calculation of the Pearson product—moment correlation coefficients
used x as the grade received and y as:

(1) the oercentage of the outside

assignments in the text read at least once; (?) the oercentage of the
library readings read at least once; (3 ) the percentage of the lecture
syllabus assignments read at least once;

(u) the percentage of the as­

signed l-'boratory guide study questions and oroblems completed; and,
(f) the number of hoars r>er week usually studied for the course exclud­
ing class time.

Pilot questionnaire findings.

None of the Pearson product-moment

r ’s computed from the raw data of the pilot questionnaires in Table I
was high ar.d only one was statistically significant according to the

2
Student t-test for the significance of r.

Frederick S. Croxton and Dudley J. Cowden, Applied General Statistics (Kew York: Prentice-Hall , Inc., l?3q )»
£*7?, £¥l.

30

TAB LI I
PILOT STUDY DATA FROM 22S STUHBNTS REPORTING ON TH3 QUESTIONNAIRES

0
^VSSTIOJTCAIRS ITEH

What ^ of the h o a ^
■work assignments in
the textbook have
r?u real r t 2--ret
-r.c e?

r
a
d
e
A
B
r
v*
D

r

All
What i of the as­
A
signed library read­- B
ings have you read
C
D
at least once?
F
All
What ^ of the lec­
ture syllabus home­
work assignment b
have you read at
least or.ee?

How nary hours per
week did you usual­
ly study for this
course excluding
class t isse?

7

2

11

17
21

?3
7

1

^5 .
l

8
29

8
2

2

Ui
H
5
15

Ui

1
5

r

3

All .... _J

A
T5

9
17
U

7
2
1
12
0
1
16
«
✓=;

2
2U

0 hrs . 3 hrs
0
5

12
2

2

0

1

0
1
1
1
0

1

2
lU
27
15

2
Q

A
B
C
E
T

9

.

33

__ 3_s_ _

0

1

10

9
?3
7

u
8

0

1

7

22

0
22

67
19

hg

U

9 hrs
1
1

21

So

37

13

25

16

7

2
1

U

2
0

All

8

nU

82

17

T
*
+J

3
o

1

68

•Statistically signi f i cant at .05 level.
••Statist! call’’ significant at .01 level.
•••Statistically si gr.i f ican t at .00 1 level •

r *»

.oil

30
U
. 139

6

6 hrs

.061

7
30

3
7

2

r —

_ ^5_.

1
6

lh

-.069

8
0

Ul

s

r —

2
lU
25

19

23

c

9

20
8

_~?3 _ . 36

B
C

Q

7

20
2

A

tc

b
07
*"—7
7

Pearson
product—
moment
correla­
tion

2

0

53

F

3

0

16
1

hg

All

What ^ of the lab­
oratory guide stu­
dy cuesticns and
r rob leers assigned
as horn.eweT.< have
ye-- -T.e?

IT-umber of students reporting in
various percentage groups the
amounta of the homework done
0-3-9 20-39 ho -89 So-79 80-100
%
*
JT
%
4>

r

** •182* •

2
125
12 hrs
0
2
1 r «* -. 05U
2
0
c

31

The hypothesis which woe drawn from these findings for further
investigation was, "There io no significant relationship between the
amount of the outside assignments which students complete and the grade
which they receive in the course when analyzed on a group ■basis,"
Certain limitations of the pilot questionnaire method were considered
"before a revised method was devised. The effecte of bias on the data
an
have been^unknown factor in all questionnaire methods of survey.
If
bias, in the form of a tendency on the part of the student to report
more homework done than was actually the case, is assumed in the pilot
questionnaire it has not been revealed in the raw data of Table I.

The

distribution of responses hAS not given any basis for assuming that stu­
dents who received the poorer grades would not tend to report deficiencies
in their performance of homework assignments.

However, if it were as­

sumed that nearly all students would not tend to reoort a lack of per­
formance, this would not be substantiated by the questionnaire data
ei ther.
Since it has been generally recognized that the bias factor is not
measurable, alternative schemes for emphasizing accurate reporting were
investigated by a trial—and— error method.
Trial revisions of the pilot questionnaire method.

Various methods

of obtaining data through questionnaires were attempted subsequently in
an effort to discover how more accurate data could be elicited from stu­
dents.

During one full terra of classes PPG students were given question­

naires each week as to the amounts of homework assignments done and the

32

number of hours studied for the week previous to reporting.

Experience

with this group suggested the possibility of supplying students with &
means of recording study performance in a diary or on some kind of
schedule.

Conversations with the students revealed that the intent to

remember and record the data as accurately as possible was frequently
frustrated by an inability to remember what they had done on the assign­
ments long enough to record the oerfornance on the questionnaire in class.
The suggestion was made to the students that the amount of the homework
done could easily be recorded in notebooks from which the data could then
be transferred to the questionnaire in class.

Few of the students to

whom thiR suggestion was made seemed able to organize a chart or plan,
or to include the recording of their study performance as part of their
study procedure.
Revision of the ^ilot questionnaire method of survey was made to
incorporate what were believed to be the best features of the various
questionnaire methods reported in the literature and of the pilot ques­
tionnaires which were tried.

It became apparent that carefully con­

structed questionnaires, requiring brief responses, administered fre­
quently to obtain information which students could readily transfer
from a carefully planned study log or diary, offered the best opportun­
ity and possibility of securing data from which valid conclusions could
be drawn.

33

D. THE CONSTRUCTION OF THE STUDY LOO AND QUESTIONNAIRE
USED IN THE INVESTIGATION
Construction of the study log.

Experience with serial question­

naires obtained during the pilot studies disclosed a need for a perma­
nent record of homework performance which the student could have in his
possession.

In order to help meet this need, and in an attempt to fa­

cilitate the reporting of data by the students, a multigraphed study
log was orepared and distributed to all students in the investigation
3
reported here.
The study log provided a concise and organized form on
which the weekly homework assignments for the entire term of classes
was recorded at the outset when the instructors gave out assignments for
the term.

Also included in the study log were columns in which the stu­

dent could make a suitable mark to indicate that (1) the assignment had
been completed, and, (?) the completion of the assignment had been re­
ported on a weekly questionnaire.

The possibility of the students* util­

ization of the study log as a study aid and self—appraisal device is
suggested in the instructions which occupy part of the page on which the
study log i8 multigraphed.

The study log was called a "progress sheet"

in the class discussions with students in an attempt to enhance the co­
operative attitude desired by the investigator.
Construction of the questionnaire.

The criteria for construction

of questionnaires which were most frequently emphasized from the survey
of the literature were:

(1) briefness of response; (?) factual nature

■^A cony of the study log may be found in Appendix II.

3^

of the information sought; and (3) Information known only to the re­
cipient of the questionnaire.

These criteria, in addition to the ex-

periencea with the pilot questionnaire, resulted in the questionnaires
used in this investigation being consistent in construction with the
study log.

Since there is no proven way of measuring the extent of the

errors in the responses, any conclusions drawn from the data must take
into consideration this unknown factor.
X. THE ADMINISTRATION OF THE STUDY LOO AND QUESTIONNAIRE
TO THE POFULATION
Admin1 3tration of the study log.

At the time of the first labor­

atory class meeting the instructors participating in the study intro­
duced the survey plan to the students and distributed study logs to
them.

As much time as was necessary was taken to answer students'

questions about the research plan and the questionnaire method which
was to be used.

A very few students expressed some skeoticisn about

being used as "guinea pigs" while several students expressed an apparent
attitude of anticipation of benefitting from the study log as a study—
guide or progress appraisal, as well as a curiosity as to their own
study performance.

Students were not informed that tie study logs would

be solicited at the end of the term's work and none of them raised, a
question in that regard.

The participating instructors did not know

that the study logs were going to be collected either, so there was no
eprarent reason for any instructor to attempt to push the students into

A copy of the weekly questionnaire may be found in Aorencix II.

35

using them.

However* from time to time these instructors were asked to

notice to what extent the students appeared to "be using the study log
to fill in the weekly questionnaires, and the usual report was that most
of the students appeared to "be keeping the study log up to date and to
be using the study log in filling in the questionnaires.

At the end of

the term students were asked to turn in the study logs and close scru­
tiny revealed that 7*3 per cent of the students had kept a detailed record
of assignments made, completed, and reported for the entire term.

Al-

thou^i the conclusions of this investigation are not based on an assump­
tion that the study log eliminated inaccurate reporting, the combination
of weekly brief—response questionnaires with the study log did seem to
minimize errors due to the memory factor to a greater extent than any
questionnaire method reported heretofore.

Administration of the questionnaire.

The participating instructors

acknowledged the possible tediousness which students might experience in
filling in questionnaires about the same thing in the same class each
week during an entire term of classes.

Students were encouraged to ask

questions and to discuss the research plan with the instructors at the
first and subsequent laboratory class meetings.

The study logs and sam­

ple questionnaires were distributed at the first class meeting as part
of the introduction of the questionnaire plan.

The operation of the

study log in relation to the anticipated weekly questionnaire was ex­
plained by the instructor at the first class meeting and as much time & b
was necessary was used in an attempt to clarify questions which students

36

asked.

The apparent tediousness of the weekly questionnaire was occa­

sionally relieved by class discussions of what might "be expected when
the data were analyzed and of the relation between performance on home­
work assignments and achievement as the students viewed the problem.
The fact that all questionnaires for the entire term were handed in
by Uo? of the U 73 students in the investigation may be interpreted as an
indication that sxiitable rapport was maintained.

However, it was recog­

nised that the effect of the instructors' use of the weekly questionnaire
to record attendance in class was no doubt reflected in the number of
questionnaires turned in.

Because of the cumulative nature of the week­

ly questionnaire, the reporting of what had been done on the assignments
since last reporting, questionnaires which were not turned in during any
particular week did not necessarily result in omission of the data from
the study.
Particir>ating staff members cooperated by reserving the first five
minutes Of each laboratory class period for filling in and collecting
the questionnaires.

Instructors were provided with the questionnaires

well in advance of the time for weekly distribution so that the ques­
tionnaire plan would require a minimum of extra work on their part.
Students were encouraged to take blank Questionnaires with them to fill
in as the homework was completed.

However, nearly all students filled

in the questionnaires during the first five minutes of the laboratory
class ooriod reserved for th°t uuroose.

Ur»on arrival at class students

were either handed the questionnaires to he completed at that time or
found conies waiting at their seats.

37

Instructors reminded the students frequently that, while the data
on the auestlonnaire absolutely would not be used in any way except for
research, attendance at class would be recorded from the questionnaires
handed in at class.

This method o^ taking roll not only motivated some

students to make a real effort to hand in the questionnaires but also
served to compensate in part for t’ e time taken fr^a classwork by the
questionnaire plan.
A complete schedule of the questionnaire plan for the entire term
was supplied to each cooperating instructor at the beginning of the term
R

of classes.

Supplementary instructions were typed and given to the in­

structors whenever the procedure deviated from the usual weekly plan.
Bores for the deposit of completed questionnaires were provided for each
instructor so that, as far as could be ascertained, no onestionnaire was
lost in handling.

As the completed questionnaires were deposited weekly

by the cooperating instructors, each questionnaire was filed with the
previous ones turned in by that student so that at the end of the term
all of the qusstiounaires turned in by any particular student had been
filed together.

Although this procedure required about an hour each

week, the laborious process of completely sorting all of the question­
naires for the entire term of classes after the term was over was elim­
inated.

Also,

the weekly sorting made possible the examination of the

completed questionnaires to determine whether or not students were com­
plying with the directions for filling them in.

In this way the impro­

perly completed questionnaires were detected early in the term and re­
sulted in an early consultation with the st.idents involved so that the
improperly completed questionnaires could be corrected.
c
A copy of the questionnaire schedule may be found in Appendix II.

3*

7.

METHODS OF ANALYSIS OF THE DATA

The number of ■variables which were to be studied and the complexity
of statistical analysis indicated the desirability of a reduction of the
amount of routine labor with the utilization of Hollerith cards.

A com­

plete flow chart of all the various operations which were neceBsary to
convert

the raw data on the questionnaires into a form which could

handled

bv IBM methods was prepared as part of the design of

Mechanical tabulation.

be

thestudy.^

Mechanical devices expedite the tabulation

of data for a statistical study when the study is extensive enough to
7
warrant a mechanical reduction of labor.
The use of tabulating: equip­
ment is recommended when there are a large number of cases involved in
an investigation for which numerous entries must be made.

The orocesa

generally consists of the following steps:
(1) Reducing all entries of the original data to a numer­
ical code.
(',) Recording these entries on a punch card (Hollerith
card) by punching holec with a key punch to represent
tne code numbers.
(5) Sorting the cards by means of an electrical or mechan­
ical sorter.
(L ) Assembling the data from the sorted cards by means of
a tabulator.

See Appendix III to examine the above flow chart.
^The mechanical devices mentioned may be leased from the Internation­
al Business Machines Corporation, 59^ Madison Ave., He-* York, N.Y.
Simi­
lar machines are available from Remington Rand Business Service, Inc.,
?1“ Fourth Ave., Hew York, N.Y.

39

Remuneration for the clerks who operated the punching, sorting,
And tabulating machines and who transferred the data from the question­

naires to the data sheets and punching schedules was provided from the
All-College Research funds of Michigan State College.
Clerical and tabulation -procedures.

The weekly sorting and filing

of the questionnaires resulted in a completely alphabetized file at the
end of the term of classes.

Since all of the questionnaires for each

respondent were filed together,

the transfer of the data from the ques­

tionnaires was thereby expedited.

An experienced clerk transferred the

data from each questionnaire onto a composite data sheet which provided
space for all entries from the questionnaires for the entire terra.

The

composite data sheet was then verified, with the clerk reading the en­
tries on the data sheet and the investigator checking the entries against the questionnaires from which the data were transferred.
clerk- then calculated, with the use of an automatic calculator,

The
the oer—

centages of the homework assignments comoleted and the average number
of hours studied as reported by the students.

Each oercentage and av­

erage was double— checked by the clerk, using the calculator.

The cal­

culated data, with the data from other sources, were transferred from
the composite data sheet to a punching schedule after assigning numer­
ical codes to the data.

The data on the punching schedule were then

verified against the composite data sheet from which the data had been
transferred and coded.

The punching of the Hollerith cards was then

done by a key punch operator from the punching schedule.

The punched

to

cards were verified "by checking a data sheet which was printed from the
punched cards.
The data for the variables which were to be analyzed were punched
S
in Hollerith cards* one for each student in the study.
Since an unknown
number of variables may have been operating to affect homework performance
and achievement, the variables of this investigation were limited to
those which previous studies have indicated might be important.
punched in the Hollerith card for each student in the study were:
1. Punch card number; each student was assigned a case number
which was punched as the punch card number to facilitate
checking the punching with the data sheets.
2. Laboratory section number in which student was registered.
3. Term grade; A- U, B- 3* C- 2, D- 1, F- 0.
U. Comprehensive examination grade; A- U, B— 3 , C- 2, D- 1,
F- 0.
5« High school quartile rank
j.st. (top) quartile— 1, 2nd
quartile— 2 , 3r<i quartile— 3 » ^th quartile- U.

6. Class of school admitted from, according to number of
students enrolled; class A (800 or more)- U, class B
(3?5-799)- 7* class C ( 150— 32^)— 2, class D (less than
150)- 1.
7. American Council Psychological Examination (Quantitative
score decile rank; 1 (lowest) through 10 (highest).

8. American Council Psychological Examination Linguistic
score decile rank; 1 (lowest) through 10 (highest).
9. American Council Psychological Examination Total score
decile rank; 1 (lowest) through 10 (highest).
10. Cooperative Test of Reading Comprehension score decile
rank; 1 (lowest) through 10 (highest).
g
See Appendix III for sample Hollerith card.

The data

Ul

11. Calculated percentage of homework assignments in the
textbook reported done.
12. Calculated percentage of homework assignments in the
lecture syllabus reported done.
13. Calculated percentage of library rending assignments
reported done.
lU. Calculated percentage of homework problems reported
done.
15. Calculated average number of hours oer week reported
studied for the course.
l£. Official student records office identification number.
The available literature describing the principles and methods of
statistics is far too extensive to summarize in this study.

Certain

citations are included in an attempt to provide examples of the typical
authoritative support which is available concerning the particular sta­
tistical methods used in this study.

Coefficients of correlation were

determined by the Pearaon-product—moment method, the generally accept­
able method of computation, an explanation of which may be found in any
recognized textbook in statistical method.

This method is found to be

so and when relationships are linear, regardless of whether or not the

9
distributions are normal.
Conimtntion of correlations has come to be an accepted method of
treating educational and psychological investigations, as indicated by
Zzekial:

^T. L. Kelley, Statistical Method (Pew York:
19-3). P. 17?.

The Macmillan Coamany,

Correlation and multiple correlation methods hare "been widely
applied in educational and psychological investigations to
the 3tudy of such problems as the relation of grades in one
subject to grades in another* or the scores on one mental test
to scores on another, or the relation of scores on mental
tests to success in the schoolroom or in later life.... In
most of the cases in which correlation analysis has been ap­
plied to psychological problems, it has been used primarily
to measure closeness of relationship rather than to obtain a
basis for estimating one variable from another.
In studies
of tills type even a low correlation may be important, so long
as it is large enough so as not to be due to random fluctua­
tions. ^
The product— moment correlation is discussed by Furfey and Daly as a re­
search techniaue and is reported to be "quite widely accepted as as
11
adequate measure of ’closeness of relationships'."
As a research
technique the product-moment correlation provides a basis for analyzing
the closeness of relationships between the variables investigated in
this study, performance on homework assignments, and the subsequent
achievement as measured by tests over the assignments.

Since the

nature of this investigation is historical rather than exoerimental,
12
according to Kelley's description of research, the simple correlation
is suitable to discover whether or not the high correlative relation­
ships occur.

"^Mordecai Ezekial* Methods of Correlation Analysis (Kew York:
Wiley and Sons, Inc., 19^7). PP« ^29» ^31.

John

1XP»ul H. Furfey and Joseph F. Daly, "Product—Moment Correlation
as a Research Technique in Education," J o u m a l of Educational Psychology.
26:206, March, 1935.
12

T. L. Kelley, Scientific Method: Its Function in Research and
Education (Columbus: The Ohio State University Press, 1929), pp. 12, US.

The variable* with the skewed distributions In the pilot study
were plotted on four scatter diagrams using random samples from the
data of the population of the four lecture sections observed in the
investigation.

The graphical evidence of the four scatter diagrams

did not refute the assumption that the line of regression of the var­
iables being studied is linear and the use of simple correlation is
suitable for this investigation.

Statistical computation.

In order to facilitate the tabulation of

the data required for computation of the Pearson product-moment corre­
lation coefficients the necessary direct ons were given to the tabulat—
13
ing machine operator on rnultigraphed U X G cards.
The front side of
the data card provided space for (l) a brief description of the varia­
bles X and Y;

(?) designation of the columns of the Hollerith cards to

be tabulated for X and Y; ("*) tabulated N (number of cards having all
ej

X and Y dAta);
values of

X, T,

(U) tabulated £x, ^Y,

p

, £y

, ,>XY; and,

rxy, and student t-test value.

filed as a oermanent record of the

data

(5) computed

These data cards are

and analysis in a reduced form.

The reverse side of the data card provided space for recording the
various values of the data when substituted in the product—moment form­
ula during the computation of the correlation for the variables in this
8 tudy.

■^See Appendix III for a sample data card.

Uh

A tabulating machine operator, using an electric sorter and com­
puter, tabulated the punched cards for the values of N, ^X, ^Y,
and ^ X Y for the variables designated by the investigator.

, £y ,

The operator

checked the machine for each sorting and tabulating.
The investigator, uring the electric automatic calculator, com­
puted the Pearson product—moment r*s from the tabulated data and checked
the computations.

The Student t— test for the statistical significance

of r was then computed for all r ’s.
G. SOME LIMITATIONS OF THE METHOD OF INVESTIGATION
Llmitatlons of questionnaire data.

Certain limitations of the log—

questionnaire method of survey in thif^stucy must be taken into consid­
eration when the findings are analyzed and interpreted.

The amounts and

effects of discrepancies tetween actual performance on the homework as­
signments and the reported performance cannot be ascertained.

The meth­

od has not eliminated the possibility of bias, either intentional or
unintentional, in reporting study nerformance.

Although the design of

the survey orcvided means by which errors in reporting could be reduced,
there is no way of determining accurate degrees of error in the data re­
ported.

Such limitations suggest that a more direct observational

method of obtaining data relevant to students*
cure more valid information.

study habits might se­

Also, a valid observational method must

he developed before a better estimate of the degree of accuracy of the
log— questionnaire method

cpr

be

obtained.

*5

Minimising errors attributable to the memory factor.

The survey

of the literature presented contradictory findings as to the reliability
of the questionnaire method of surveying students.

No reports were

found which purported to investigate the validity of the questionnaire
method other than by computing a coefficient of reliability as a measure
of validity.

Close scrutiny of the methods which have been reported as

questionnaire methods disclosed the fact that respondents were required
to recall the information asked for in the questionnaire.

In the typical

survey the respondents have not been given practice in answering the type
of cuest.ions making up the survey.

None of the methods examined gave

respondents to questionnaires asking for attitudes and opinions adequate
instiuctions and time for reflection.

Reasonable guesses have been made

to explain very high or very low reliability coefficients for such survey
methods, but no variation of the questionnaire method was found which
would detect the effects of the various attempts to inrorcve the method of
administering the questionnaires.
The log-questionr.nire method of survey used in thie investigation
makes nossihle the reduction of dependency upon memory to a minimum.
Every oossihility for maintaining high rapport with the students was
exploited.

Students were supplied with details of the homework assign­

ments for the entire term at the first laboratory claes meeting.

They

were encouraged by instructors to utilice the study log as a study aid
and to record the completion of the assignments for an accurate report­
ing on the weekly questionnaires.

Although the discrepancy between ac­

tual performance on the assignments and the reported performance has not

U6

■been ascertained, the mechanics of the log-questionnaire method of sur­
vey "alee more accurate resronsps possible.

This method did not require

respondents to fill in the questionnaire from memory but encouraged, by
virtu® of the permissive nature of the plan,

the recording of the date,

from the students* own records on the study logs.

Since it was found

that 78 per cent of the respondents had kept a complete record of per­
formance on homework assignments, it may be inferred that the log-ques—
t-ionnaire method of survey reduced the effects of errors due to memory
to a minimum.

Correlation coefficients frcn the re-teke data.

Two estimates of

the students* ability to copy the data from the study log to the ques­
tionnaire were comrmted.

One estimate was obtained from a written re­

take cf the questionnaire.

The second estimate was obtained from an

oral re— take of the weekly questionnaire.

The written re— teke consisted

o' having all the students in the study, in attendance during the class
reriod when the re— take was river., fill in a second cory of the regular
weekly questionnaire.

Th*» re-take was administered during the last few

minutes of a laboratory period with the instructions that the students
were to attenrot to duplicate the regular questionnaire which had already
been turned in at the beginning of the chess period.

The students were

permitted to use stud;' logs or any other means which they chose to use
in filling ir. the re-take questionnaires.

This re-take was given un­

announced to all respondents during a regular class period near the mid­
dle cf the term of classes.

*7

The data and calculated Pearson product—moment r*s obtained from
the regular weekly questionnaire responses and the written re— take re­
sponses have been included in Table II.

The data of Table II were ob­

tained from the original and re-take questionnaires given during a
single Period with items I, II, and III being the number of pages of
reading assignments reported done, item IV the number of assigned prob­
lems done, and item V the number of hours reported studied for the
course.

The X variable, data from the regular weekly questionnaire, and

the T variable, data from the written re— take, were then substituted in
the simple correlation formula.
The estimates of reliability of the students* ability to transfer
data from the study logs to the questionnaires were found to be .9^7 or
higher.
A second estimate of the reliability of the students* ability to
transfer data from the study logs to the questionnaires was obtained
from an oral re-take questionnaire of a random sample of 25 of the re­
spondents in this study.

The oral re-take questionnaire data was se­

cured by personal interviews near the end of a regular laboratory class
period in which the students had already turned in the usual weekly
reports.

Students making up the random sample subjected to the oral

re-take questionnaire, were permitted to refer to study logs or any
other means available for answering the questions, which were identical
with those of the weekly questionnaires.

The oral re-take was given un­

announced to the random sample near the middle of the term.

The data and

the calculated Pearson product-moment r's secured from the oral re-take
reports may be examined in Table III.

l+S

TABLE II
DATA FOR CALCULATION AND THE CALCULATED SIMPLE CORRELATIONS OF THE QUES­
TIONNAIRE ITEMS FROM ORIGINAL AND WRITTEN RE-TAKE
Questionnaire data reported by students
Item II
Item III
I tern I
Item V
Item IV
pp. text
pp.
library
No.
probe.
hrs.
stud.
pp.? srii.-.
2966

3U 16

1385

1885

761

?983

3U66

1U?5

1872

761

(*p

67913

91256

88U29

2 9 956

2U50

t?2

665S7

923 80

^9057

28830

2UU6

91375

88371

28898

2U3 I

Up s

ups

Up 5

U25

£

(Original
questionnaire)
(Re-take
questionnaire)

*

ixr

N
r

ups
•9U7**

• 993**

•99s**

•Statis^pally significant at .05 level.
••Statistically significant at .01 level.
••• Statistically significant at .001 level.

• 997**

.985**

table

h i

DATA FO R CALCULATION AND THE CALCULATED SIMPLE CORRELATIONS OF THE QUES­
TIONNAIRE ITEMS FROM ORIGINAL AND ORAL RE-TAKE

Que s t ionnai re data reported by students
Item III
Item I
Item II
Item IV
Item V
T>p• toxt p p . sy 1 1 . pp. library Nc. -nrobs. hrs. stud.

(x (Original
questionnaire)

165

201

1 L7

38

51

£y (Oral re-take
cmestionnaire)

150

203

172

LO

51

2913

3625

5155

200

175

269^

3661

5812

216

175

269^

3 6 L1

5117

206

175

N

25

25

25

25

25

r

.9^**

•99**

£x y

.9 2 **

•Statistically significant at .05 level.
♦•Statistically significant at . 0 1 level.
•••Statistically significant at . 0 0 1 level.

.98**

1 .0 0 **

50

The data of Table III were obtained, from the regular and oral re­
take questionnaires given during a single class period with items I, II,
and III "being the number of pages of reading assignments reported done,
item IV the number of assigned problems done, and item V the number of
hours reported studied for the course.

The X variable, data from the

regular weekly report, and the Y variable, data from the oral re— take,
were then substituted in the correlation formula.

The estimates of re­

liability of the students* ability to copy data from the Btudy logs to
the questionnaires were found to be .9? or higher by the oral re— take ques­
tionnaires administered to a random sample of the respondents in this
study.
Interpretations from simple correlations.

The use of simple corre­

lations in this study will result in very limited interpretations of the
findings.

The correlations will not show to what extent some unknown

variables may be more related to the problem than those analyzed.

Sta­

tistical correlations do have a definite value in research in that:
do not discover causal connections by first surveying all
possible correlations between different variables.
On the
contrary, we suspect an invariable connection, and then use
correlations as corroborative evidence.... The method of
concomitant variation cannot therefore be accepted as a meth­
od of either discovery or proof.
Its value lies partly in
suggesting lines of inquiry for causal relations and in help—
inf* to corroborate hypotheses of causal connection.
Its chief
value, however, is to help eliminate irrelevant circumstances
For nothing will be regarded as the cause of a phenomenon if
when the phenomenon varies that thing, does not, or when the
phenomenon does not, that thing does. ^

'Se

Morris R. Cohen and Ernest Nagel, An Introduction to Logic and
Scientific Lethod ( N e w York:
Harcourt, Brace and Company, ld^u^, pp.
W f - T U i . -----------

51

Providing the requirements of the statistical method have teen met
in the investigation, the correlation supplies probability statements
concerning the correctness of the inferences which may be drawn from
the calculated coefficient.

52

CHAPTER. IV
PRESENTATION AND ANALYSIS OF THE DATA

A. DATA AND STATISTICAL INFERENCES
The data of Table IV indicate the correlative relationships between
the reported homework performance and the achievement of the students as
determined by the term f m d e assigned b;.” the instructors on the basiB of
tests given over the assignments.

If there Is a causal relationshio be­

tween homework performance and achievement, which is the assumption under­
lying the typical homework assignment investigated here, a high correla­
tive relationship may be ascertainable.

Even though very high correla­

tions are not necessarily indices of causal relationships, any assertion
of cause-effect relationship must be corroborated by high correlations.
In all tables of data pertaining to homework performance the N
(number of cases) for performance of library reading assignments is less
than for tVe other homework r>erformance variables because one of the
'o u t instructors participating in the study did not assign library read­
ings .
The data of Tables IV, V, VI, VII, VIII, IX, X, XI, XII, and XIII
were substituted in the simple correlation formula for the conroutation
of the r* s shown in the tefbles.

In each coro^utation of the r ’s, the

deta in the t^o section of each table were used.

53

Essentially, the datn of Table IV ma;/ he analyzed to test the a s ­
sumed positive correlative relation between homework; performance and
achievement investigated in this study.

The low correlation,

.105* be­

tween the percentages of the assignments in the textbook reported read
an:, the tern grades received should not be considered as corroboration
of the assumed relationship.

The otner low correlations,

(1) .0U 5 be­

tween term grades and percentages of the lecture syllabus assignments
reported read;

(?) .09b between term grades and percentages of problems

reoorted done; and (3) — » 0 J 1 between terra grades and average number of
hours per wee-: reported studied for the course, are indicative of the
lack of the commonly assumed cause— effect relationship.

The higher

correlation of .38b between term grades and the percentages of library
readings reported road may indicate some emphasis on library readings in
the tests w' ich determined the term grades.

While the low correlations

do not support the assumption being tested, neitner do the low corrola—
tions explain why the expected relationship appears to be non-existent.
An inference which may be drawn from the data is that the factors which
contribute to or cause the term grades apparently are not inherent in
the homework performance data as reported by the students.
Table V includes the correlations which were computed between terra
trades and other variables which previous investigations have suggested
as possible contributors to academic performance.

Such analysis reveals

correlations which have high statistical signifinance:

(1)

,?01 be­

tween term grades and, Cooperative Test of Reading Comorehension decile
rank;

(?)

.lb? between terra grades and American Council Psychological

5*

table

iv

DATA FOR CALCULATION OF SIMPLE CORRELATIONS OF HOMEWORK PERFORMANCE
WITH TERM GRADES

Questionnaire Data
Per cent
of text
reported
read

(term grades)
(study habits)

if

Per cent
of lec­
ture
syllabus
reported
read

3 ?5 « 0

925.0

25363.U

Per cent
of li­
brary
readings
reported
read

Per cent
of prob­
lems
reported
done

Average
number of
hours per
weelc re­
ported
studted for
course

697.0

925.0

925.O

1U71S.3

1 U9 U6 .U

13U7.5

1 6 U9 . 0

2 1 7 0 .0

2195.O

2 1 9 5 .0

2 1 9 5 .O

if

lgt+3733.0

2569555.5

1161662. s

85U62U.g

6250.5

£x y

51710.7

62601.0

30S97-9

30713.3

260b.6

N

U6 6

U56

3US

U6 6

.1 0 5 *

.0 U 5

•Significant at .05 level.
**Signifleant at .01 level.
***Significant at .001 level.

,3 SU***

.0 9 U*

U66
-.0 7 1

55

TABLE V
DATA FOR CALCULATION OF SI MPLS CORRELATIONS BETWEEN TERM GRADES, CERTAIN
INDICES OF GENERAL COLLEGE ABILITY (COOPERATIVE TEST OF READING COMPRE­
HENSION DECILE RANK, AMERICAN COUNCIL PSYCHOLOGICAL EXAMINATION Q, L, AND
TOTAL SCORE DECILE RANKS), AND CLASS OF HIGH SCHOOL FROM WHICH GRADUATED
Indices of General College Ability and Class of High
School from which Graduated
Coopera­
American
American American Class of
Council
Council
Council
high school
tive Test
Psycho­
Psycho­
Psycho­
from which
of Read­
logical
ing Com­
logical
logical
graduated
Examina­ Examina­
prehension Examina­
tion Q,
tion To­
tion L
decile
rank
score
tal score
score
decile
decile
decile
T-nnV-----rwr>lr____
rank— —

3 9 1 .0

3 9 1 .0

3 9 1 .0

3 9 1 .0

620.0

^Y (indices)

2U30.O

2U79.O

?Ul?.0

p L?U.O

9 5 1 .0

¥

2103.0

2103.0

2103.0

2103.0

1 U6 3 . 0

16336.0

16730.0

1615?.0

161H2.0

3 1 9 3 .0

5030.0

5013.0

U9U6.O

U96I.O

1 3 9 5 .0

(term grade)

(XY

LL9

N

xy

.201 ***

ULg

.092*

UU9

.1U2***

•Statistically Significant at .05 level.
••Statistically significant at .01 level.
•••Statistically significant at .001 level.

UU9

.lhg***

312
.019

56

Examination Linguistic score decile rank; and (3) *1^9 'between term
grades and American Council Psychological Examination total score de­
cile rank.

Such low correlations cannot be considered as corroborative

evidence of the assumed relationships.

The correlations of (l) . 0 9 2

between the term grades and American Council Psychological Examination
Quantitative score decile rank; and (?) .019 between term grades and
class of high school from which graduated are so low that practically
no relationship can be assumed for these variables.

Although low cor­

relations are revealed for the additional indices of academic success,
the inference can be made that perhaps some combination of factors
would produce a higher correlative relationship.

The correlations be­

tween term grades and the psychological and reading test decile ranks
are large enough, however,

to permit the inference that apparently the

variables measured by such tests have a higher relation with the tests
whereby the term grades were assigned than with the homework perform­
ance reported by the students.
Asterisks (•) are to be found in all tables in this study where the
r is significant statistically.

One asterisk next to an r (*) indicates

that the level of significance of the r is at five per cent (5/6); two
asterisks next to an r (**) indicate that the level of significance of
the r is at one per cent (1$); three asterisks next to an r (***) indi­
cate that the level of significance is at one— tenth of one per cent
(.0 0 1 $).

Levels of significance were determined by consulting the t

table of Crouton and Cowden^and referring to the values of t and n
^Frederick E. Croxton and Dudley J. Cowden, Applied General Sta­
tistics (New York: Prentice-Hall, Inc., 1 9 3 9 )• P* 3 7 5 *

57

(n - N — ?)•

The t table, upon proper application of t and n, indicate*

"how many times in 100 a sample drawn from a population with zero cor­
relation would result in a correlation coefficient as high as that ac­
tually obtained.

If this chance is very low, the correlation ie assumed

to be significant."^*

If, upon application of the t— test for the signifi-

cance of an r, the t table indicates that there would be five (5) or less
chances in 100 that a sample drawn from a population with zero correla­
tion would result in a correlation coefficient as high as the r actually
obtained, the r is termed "statistically significant."
Some correlations, even though statistically significant, are too
small for a reasonable prognostic value:
Relation of the Correlation Coefficient to the
Per cent of Forecasting Efficiency
r
O
H
•

.20
.50
.60
.70
.30
-90
.95
•93
1.00

E

.5
2.0
5.0
8.0
13.0
20.0
?9 .o
Uo.o
56.0
69.0
80.0
100.0

Croxton and Cowden, op. clt., p. 6 3 1 .
^Clark L. Hull, "The Correlation Coefficient and its Prognostic
Value," Jouraal of Educational Research, 15*335» May, I927.

58

Holzinger offers the following interpretation of correlation
coef ficients:
As an example* it may "be noted that a correlation coefficient
of .SO gives a value of l_m UO £Note that S • Uo according to
HullJ , which means that the regression forecast with a single
score is here only Ho per cent better than a random guess.
Analysis of the data of Table VI shows that the assumed correlative
relationships between the comprehensive examination grades and the reoorted homework performance are not corroborated by this study.

Such

low correlations of the conorehensive examination grades. — .059 with per­
centages of lecture syllabus assignments reported read.

.Q?l with the

percentages of textbook assignments reported read, .063 with percentages
of library readings reported read, — .163 with percentages of assigned
prob-leins reported done, and .0T6

with theaverage number of hours per

week reported studied for the course, may be due to the fact that the
homework performance was reported

for only one tern while the conorehen-

sive examination covers the three

terms ofthe coarse.

When the comprehensive examination trades are studied in relation­
ship to tv e same additional indices of academic achievement as were the
term grades,

certain differences are very apparent.

Of all the factors

analyzed which might contribute to the comnreher.sive examination grades,
the psychological and reading test deciles were the best indicators ac­
cording to the correlations of the comprehensive examination grades,

.^37

with the Cooperative Test of Reading Conpre’ enaion decile rank, .?Ul with

~ \ a r l J. Holzinger, Stn ti s tical Methods for Student s in Education
(Boston:
Ginn and Company, lh?*7). p . 167•

59

the American Council Psychological Examination Quantitative score decile
rank, *379 with the American Council Psychological Examination Linguistic
score decile rank, and *^13 with the American Council Psychological Exam­
ination (hereafter, abbreviated to ACE in this report) Total score decile
rank.

Apparently the comprehensive examination grades have a higher rela­

tion with the psychological and reading test deciles than do the term
grades.

The correlations (l) of .1+37 between the comprehensive examina­

tion grades and the reading test decile rank, and (?) of «379 between the
comprehensive examination grades and the ACE Linguistic score decile rank,
reveal a definite trend for the comprehensive examination grade to be an
indicator of reading and linguistic ability. The inference car. be made that
the comprehensive examination is more of a reading end linguistic ability
test, than the term tests to the extent that the higher correlations in­
dicate hi .‘rhea rf'lat ionsh ins between performance on the conrnrehensive ex­
amination and reading and linguistic deciles than between the term teste
and the reading and linguistic deciles.

The higher correlations between

the reading and linguistic deciles and the comprehensive examination
grades, when compared with the correla-tione between comprehensive exam­
ination grades and the Quantitative score deciles, indicate that the
comprehensive examination is more of a measure of reading and linguistic
ability than of quantitative ability which is purported to be more sig­
nificant for the prediction of success in scientific and technical cur­
ricula than reading and linguistic ability as measured by the ACE tests.
Since the comnrehersive examination for the Biological

Science course is

constructed to measure student achievement in terms of the stated course

6o

objectives an

inference can Ve made thet the coursework, as veil as

the examination of the coursework, io more dependent upon reading and
linguistic ability than upon quantitative ability.
Although the correlations of Tables VI and VII do not prove that
certain skills, particularly the reading and linguistic skills, are the
principal variables for determining grades received on the comprehensive
examination, the conclusion that the data do not sur.port the assumed
cause— effect relationships between homework performance and subsequent
performance on tests over the homework,

seems reasonable.

Since the principal problem under investigation concerns the as­
sumed positive correlative relation between homework performance and
achievenprd, analysis of other variables is the next step.

The quanti­

tative analysis of achievement on the tests piven by instructors as
represented by toria grades and the achievement on the comprehensive
examination as represented by the comprehensive grades deals with the
quantitative aspects of the performances on the homework assignments as
reported by the students.

No data is available from which qualities of

the instructors1 tests can be ascertained except the instructors*

re­

ports that they attempted to cover the assignments in the tests t'< ey
gave to students.

The limited information concerrinr the qualities of

the comprehensive examinations seems to indicate that the comprehensive
examinations do a better Job of examining students over the coursework
because of the way test cutetruetion criteria are used to construct and
5
evaluate the examinations.
^Board of Examiners, Michigan State College, Comprehenaive Examlna—
t ions in a_ Program of General Educat ion (East Lansing:
Michigan State
College Pr^ss, 19 ^ 9 7 7 ”pp.
--------

61

TABLE VI
DATA FOR CALCULATION OF SIMPLE CORRELATIONS OF HOMEWORK PERFORMANCE
WITH COMPREHENSIVE EXAMINATION GRADES
Questionnaire Data
Per cent
of text
reported
read

Per cent
of lec—
ture
syllabus
reported
read

Per cent
of 11—
brary
readings
reported
read

Per cent
of problose
reported
done

^!C (comp, grades)

1007.0

1007*0

762.0

1007.0

1007.0

(y (study habits)

2 ^2 0 0 . 7

3 0 9 2 8 .0

IU77I.O

IU70U.3

1 2 9 5 .0

I*

?53i.o

2518.C

1923.0

2531.0

2531.0

if

ISH37S3.I

2U38525.9

1168015.9

9U6ggQ.U

61*59.^

£XY

55009.6

6676U.I

33003.3

2 9 9 8 1 .5

2951.0

UUc

U(;2

3 i49

U62

UU 5

.021

-.059

.063

♦Statistically significant at .05 level.
♦♦Statistically significant at .01 level.
♦♦♦Statistically significant at .001 level.

-.163***

Average
number of
hours per
week re­
ported
studied for
course

.026

62

TABLE VII
DATA FOR CALCULATION OF SIMPLE CORRELATIONS BETWEEN COMPREHENSIVE EXAM­
INATION GRADES AND COOPERATIVE TEST OF READING COMPREHENSION DECILE RANK,
AMERICAN COUNCIL PSYCHOLOGICAL EXAMINATION Q, L, AND TOTAL SCORE DECILE RANK,
AND CLASS OF HIGH SCHOOL FROM WHICH GRADUATED

Cooperative Test
of Reading
Comprehension
decile
rank

American
Council
Psychological Exam,
Quanti tative score
decile
rank

Questionnaire Data
American American
Council
Council
Psycholog­ Psychology
ical Exam. ical Exam
Linguis­
Total
tic score
score
decile
decile
rank
rank

Class of
high schot
from whici
graduated

976.0

976.0

976.0

976.0

677.0

^ Y (indices)

2375-0

2 4 0 5 .0

2 3 5 1 .0

2360.0

9 0 2 .C

&

244s. 0

2448.C

2448.0

244s. 0

1727.0

16092.0

1 6 6 0 0 .c

1 5 9 2 2 .0

15S6 9 .0

3164.0

5741.0

564s. 0

5 6 5 s. 0

56S3.O

^ X (comp, grades)

(X*
N

rxy

43 2

.4 3 7 ***

432

432

.3 7 9 ***

432

.4l3***

2 0 5 2 .0
296

-.040

•Statistically significant at . 0 5 level.
••Statistically significant at . 0 1 level.
•••Statistically significant at . 0 0 1 level.

i

63

The data of Tables VIII, IX, X, XI, XII and XIII pertain to some
of the variables which may be affecting homework performance as re­
ported by students.

The variables studied by means of computing sim­

ple correlation coefficients are:

1 . Cooperative Test of Reading Comprehension decile ranks.
?. American Council Psychological Examination Quantitative score
decile ranks.
3 . American Council Psychological Examination Linguistic score
decile ranks.
H. American Council Psychological Examination Total score decile
ranks.
5 . Quartile rank in high school graduation class.
6. Class of hi^a school from which graduated.
The correlation coefficients of Tables VIII, IX, X, XI, XII, and
XIII are the bases for a series of interpretations concerning the re­
lations between the various aspects of homework performance reported by
students and the variables listed above.

The negative and very low

correlations between the percentage of the textbook assignments reuorted
read and the above indices indicate that no relationship exists between
them.

In particular,

the amount of the outside assignments in the text­

book re-oorted to have been read by the students bears no relationship
to rending ability, general college ability, quartile rank in high
school graduation class, and the class of high school from which grad­
uated.
The data also support the contention that linguistic ability may
not be associated to any extent with the amount that is actually read.
The very low arid negative correlations found between the percentage of
the homework assignments in the lecture syllabus reported read and the
indices analyzed indicate little or no apparent relationship.

Reading

6h

TABLE VIII
B A T A FOR. C A L C U L A T I O N OF S I M P L E C O R R E L A T I O N S O F H O M E W O R K P E R F O R M A N C E
W IT H D E C I L E RAN K A S D E T E R M I N E D B Y T H E C O O P E R A T I V E T E S T O F R E A D ­
ING C O M P R E H E N S I O N S C ORES

Per cent
of text
reported
read

(reading; dec.)

Per cent
of lec­
ture
syllabus
reported
read

Questionnaire Data
Per cent Per cent
of li­
of prob­
brary
lems re­
readings ported
reported done
read

2UU5.C

1805.0

2^5.0

Average
number
of hours
per week
reported
studied
for couri

^Y (study habits)

2U7U8.7

3 0 ^5 5 . 2

IU313.6

1U225.1

132s.6

<?

1 6 3 7 6 .0

I6U05.O

120U1.0

I6U05.O

16U05.0

1 8 3 1 6 0 2 .7

2 5 1 1 5 9 0 .8

1 1 3 5 6 7 6 .1

912UU5.9

6227.7

^XT

I300L.5

159218.U

7 5 UUU . 3

7U003.8

76U5.2

N

U53

L53

335

U53

U53

-.0^5

-.072

-.090

13 U**

•Statistically sigrificant at *05 level.
••Statistically significant at .01 level.
•••Statistically significant at .001 level.

•173***

65

TABLE IX
BATA F O R CALCULATION OF SIMPLE CORRELATIONS OF HOMEWORK PERFORMANCE
WITH DECILE RANK AS DETERMINED BY THE AMERICAN
COUNCIL PSYCHOLOGICAL EXAMINATION Q, SCORE

Per cent
of text
reported
read

Per cent
of lec­
ture
syllabus
reported
read

questionnaire Data
Per cent Per cent
of li­
of prob­
brary
lems re­
readings ported
reported done
read

Average
number
of hours
per week
reported
studied
for the
COUT8S

?l+gg.D

21+88.0

I8L3.O

?Lgs.o

?l+gg.O

?U 7 U s . 7

30U55.?

IU3I3.6

ll+??5 .l

1328.6

169??.o

169??.0

1?605.0

169??.0

169??.0

i f

1&316o ?.7

?5 1 1 5 9 0 .8

1 1 3 5 6 7 6 .1

9 l?l+i+5 . 9

6 ?? 7 . 7

^XY

13 ? 3 3 6 . 7

163663.1

806?U.1+

7176U. 5

7 6 1 9 .9

33k

1+53

1+53

^X Cq decile)
^Y (study habits)

N
r

1+53

XZr

-.091

U53

-.093*

.01+6

♦Statistically significant at .05 level.
•♦Statistically significant at .01 level.
•♦•Statistically significant at .001 level.

.1 1 7 *

66

TABLE X
DATA FOR CALCULATION OF SIMPLE CORRELATIONS OF HOMEWORK PERFORMANCE
WITH DECILE RANK AS DETERMINED BY THE AMERICAN
COUNCIL PSYCHOLOGICAL EXAMINATION L SCORE
Questionnaire Data
Per cent
of text
repo rt ed
read

Per cent
of lec­
ture
sjtllabus
reported
read

Per cent
of li­
brary
readings
reported
read

Per cent
of prob­
lems
reported
done

zhzo.o

2U20.0

1776.0

2U20.0

2^20.0

2U7US.7

30^55.2

IU313.6

1 U?P 5 . 1

1 3 2 8 .6

16170.0

16170.0

1 1 8 2 2 .0

16170.0

16170.0

1831602.8

2511590.8

1 1 3 5 6 7 6 .1

9 1 2 LL5 . 9

6227.7

(XT

130U15.5

1 5 9 3 3 2.6

7 M 5 1 .7

7 3 ^8 9 - 0

7555*8

N

^53

335

^53

-.0U9

-.06U

(x (Ujlecile)
^Y (study habits)

&

r

xy

-.01+6

1+53

-.0 8 7

♦ S t a t i c a l l y significant at .05 level.
•♦Statistically significant at .01 level.
* ♦♦Statistically significant at .001 level.

Average
number
of hours
per week
reported
studied
for the
course

U 53

67

TABLE XI
DATA FOR CALCULATION OF SIMPLE CORRELATIONS BETWEEN HOMEWORK PERFORMANCE
AND THE AMERICAN COUNCIL PSYCHOLOGICAL EXAMINATION
TOTAL SCORE DECILE RANK

Per cent
of text
reported
read

iX (ACE Total
decile)
£y (st illy habits)

iy
ixy
N

2U31.0

pim.o

Average
number
of hours
per week
reported
studied
for the
course

1 7 9 1 .0

0^31.0

?U3i.o

?U 7 U8 . 7

30U53.2

IU3I3.6

1UP23.1

1328.6

161PG.0

1615?.0

11833.0

16155.0

16155.0

IS31602.7

2511590.8

1135676.1

1 2 9 7 7 8 .5

159863.U
U 53

U53

TV

Per cent
of lec­
ture
syllabus
reported
read

Questionnaire Data
Per cent Per cent
of li­
of prob­
brary
lems
readings reported
reported done
read

.079

-.09V

n- -•

:.h

6227.7

75766.3

71970.7

7 6 3 9 .0

335

U53

U53

-.022

♦Statistically significant at .05 level.
♦♦Statistically significant at .01 level.
♦♦♦Statistically significant at .001 level.

- . 115*

.1 8 9 * **

table
data fo r

calculation

of

simple

xii

correlations

of homework

performance

WITH QUART ILE RANK IN HIGH SCHOOL GRADUATION CLASS

Per cent
of text
reported
read

Questionnaire Data
Per cent
Per cent Per cent
of lec­
of li­
of prob­
ture
brary
lems
syllabus
readings reported
reported
reported done
read
read

Average
number
of hours
per week
reported
studied
for the
course

64s. 0

64s.0

477.0

64s.0

64?.o

19764.9

24373.5

120?4. 9

11652.3

1 0 2 6 .7

l44?.0

1448.0

1046.0

1428.0

1779.0

1 4 6 1 1 5 0 .1

2 OOO6 S8 . 5

952250.^

729374.0

3107.0

£XY

33977.9

4P309.9

2 0 1 0 7 .3

19716.0

1498.4

N

365

365

?7?

368

-.09?

-.104

(H.S. quart.)
^Y (study habits)

if

r XV

-•1 0 3 *

♦Statistically significant at .05 level.
♦♦Statistically significant at . 0 1 level.
♦♦♦Statistically significant at . 0 0 1 level.

-.0 7

s

36 5

-,874^'

69

TABLB XIII
DATA FOR CALCULATION OF SII^PLE CORRELATIONS OF HOMEWORK PERFORMANCE
WITH CLASS OF HIOH SCHOOL FROM WHICH GRADUATED
Qvieationnaire Data
Per cent
of text
reported
read

Per cent
of lec­
ture
Syllabus
reported
read

Per cent
of 11brary
readings
reported
read

Per cent
of problema
reported
done

9 5 ? .0

9 5 ^ .0

693.0

95?.0

9 5 ?-0

17053-3

P091S.9

IOO7 6 . 7

IOIU7 .I

35?.h

3i?6.o

319U.O

2279*0

8630.0

319U.O

l?6 7 7 3 7 . 6

l7 ??hoU.g

7 9 5 ?'^o.?

7 ?9 3 7 U.O

3669.8

£XY

51I89.9

63353 .h

31026.7

U5919.9

?6^6.7

N

313

313

53 l

315

313

-.077

.075

.09L

,313

.027

(class of R.S.)

{Y (study habits)

• S W . i s t o n i l y significant at .05 level.
••Statistically significant at .01 level.
•••Statistically significant at .001 level.

Average
number
of ho lira
per week
reported
studied
for the
course

70

ability* general college ability, quartile rank in high school gradua­
tion class, and class of high school from which graduated bear no im­
portant relation to the amount of the outside assignments in the lecture
syllabus reported done by students,
Armarently performance on the library reading assignments as report­
ed by the students bears no relation to reading ability, general college
ability, quartile rank in high school graduation clas 3 , and class of
high school from which graduated.
Some secondary school authorities have suggested that in general,
the larger schools more commonly give more homework problems to stu­
dents than do the smaller schools, perhaps because of pressure of larger
classes.

The correlation coefficient of *313 between the percentage of

the assigned problems reported done by students and the class of high
school from which graduated may be interpreted as a possible effect of
this reported practice.

However, the *313 correlation is still too

small to be of nruch predictive value as far as homework performance is
concerned.

Such a correlation does indicate a higherelation between

nerf^rnance on the assigned problems as renortad b y students and the
class of high school from which graduated than between performance on
ossifrned problems and any of the other indices analyzed.
The amount of time utilized by students for preparation for a t>articulnr course of study has been the subject of other investigations
but the reoorted results are contradictory.
veal no significant correlative

The data of this study re­

relationships between

the amount

of

71

time reported studied for the course and the subsequent achievement in
the course as measured "by term grades or comprehensive examination
grades.

Statistically significant correlative relationships were found

"between the average number of hours per week reported by students as
studied for the course and the indices analyzed, with the exception of
the correlation of .0?7 with the class of high school from which grad­
uated .
One high correlation was obtained from the analysis of the varia­
bles which might affect homework performance.

The relation between tne

academic achievement of the students in high school, as indicated by
ouArtile rank in the graduation class (from top to bottom), and the
amount of time reported studied for the Biological Science course was
found to be .8 7 ^»

Such a nigh correlation, by comparison with the

others derived in. this stud:/, is difficult to explain.

For even if the

assumption were made that the higher ouartile rank in high school
achievement is related to a higher number of hours studied for the high
school courses, and that the patterm of the number of hours studied is
carried over to college work,

the expectation that the number of hours

studied for college courses would be related to college achievement is
not realized according to the data of thi6 study.

There is no apparent

reason why quartile rank has such a high relation to the number of hours
reoorted studied for the course,
A more explicit conclusion from tne data pertaining to quartile
rank would state that, while a high correlative relationship was found
between quartile rank in high school graduation class and the average

72

number of hours per week students reported they Btudied for the Bio­
logical Science course, no apparent relationship was found between the
number of hours reported studied and the subsequent achievement as de­
termined by term grades and comprehensive examination grades.
The data of this study corroborate the findings of several inves­
tigations which have reported high school standing to have the highest
prognostic value for college success of all factors which have been
studied in the correlations of .3^3 between high school quartile rank
and term grades in the course and .?2b with the comprehensive exaaina^tion grade.

Although these are not as high as some of the reported

correlations they are statistically significant.
To apparent relationship was found between the amount of time re­
ported studied for the course and the size of high school from which
the student graduated.
B.

SUMMARY

OF

STATISTICAL INFERENCES

The correlation coefficients which were coranuted, analyzed, and
Interpreted for this etudy, with the exception of tnose reported as
pilot study findings and measures of questionnaire reliability, have
been arranged in summary in Tables XIV and XV.

Certain statistical

inferences may be drawn from the correlation data computed for the anal­
ysis of achievement in this study.

A positive and statistically signi­

ficant correlation was found between achievement, as measured by the
term grades, and (1) percentage of homework assignments in the textbook
reported read; (P) percentage of assignments in library readings reported

73

TABLE XIV
SUMMARY OF CORRELATION COEFFICIENTS COMPUTED
FOR THE ANALYSIS OF ACHIEVEMENT

Achievement
Vnrin'blftn analysed
Term grade

Percentage of assignments
in textbook reported read

.1 0 5 *

Percentage of assignments
in lecture syllabus
reported read

.0 U5

Percentage of assignments
in library readings
reported read

•3SU***

Percentage of problems
reported done

.0 9 U*

Average number of hours
per week reported studied
for the course

- .0 7 1

Comprehensive exam­
ination grade

.0 2 1

-.059

.063
•.163***

.0 2 6

Cooperative Test of
Reading Comprehension
decile rank

.2 0 1 **•

.k3J***

ACE Quantitative score
decile rank

.0 9 2 *

.2U3 ***

ACE Linguistic score
decile rank

.lU2***

•379***

ACE Total score decile rank

.Iks***

.Ui3***

Quartile rank in high
school graduation class

•3^3***

.226***

Class of high school
from which graduated
.019
•Statistically significant at .05 level.
••Statistically significant at .01 level.
••• Statistically significant at .001 level.

-.0 U0

TABLE XV
SUMMARY OF CORRELATION COEFFICIENTS COMPUTED FOR THE
ANALYSIS OF HOMEWORK PERFORMANCE

Per cent
of text
reported
read

Term grade

.1 0 5 *

Comp.exam, grade

.0 2 1

Per cent
of lec­
ture
syllabus
reported
read

.0 U5
-.0 5 9

Homework Performance
Average
Per cent Per cent
of li­
of -prob­
number
brary
of hours
lems
per week
readings
reported
reported done
reported
read
studied
for the
course
.3 SU***
.0 6 3

.0 9 H*

- .0 7 1

- . 1 6 3 ***

.0 2 6

-.189***

.1 1 7 *

ACE Quantitative
score decile rank

-.091

-.093*

ACE Linguistic
score decile rank

-.0 U 6

-.027

-.0 U9

-.06U

.166***

decile rank

-.079

-.0 9U

-.0 2 ?

-.115*

.1 3 9 ***

Cooperative Test
of Reading Comprehen eior. doc. r-..me

-.0 9 0

-.I3 L * *

-.0 ^ 5

-.0 7 ?

.173***

Quartile rank in
high school grad­
uation class A

-.1 0 3 *

-.09?

-.10U

-.0 7 *

- . 8 7 U***

Class of high
school from which
graduated

-.077

.075

.09U

.0 ^ 6

ACE Total sod re

.313***

.027

The negative r's for the quartile rank with these same variables,
as shown in Table XII, were negative because of the coding; the
relationships were positive, however, since the highest ouartile
ran> was coded 1 and the lowest U.
•Statistically significant at .05 level*
••Stntistirally significant at .01 level.
***Statistically significant at .001 level.

75

read; (3 ) percentage of homework problems reported done; (U) Coopera­
tive Test of Reading Comprehension decile rank; (5 ) ACE Quantitative
score decile rank; (6) ACE Linguistic score decile rank;

(7 ) ACE Total

score decile rank; and (?) quartile rank in high school graduation
class.

Positive and statistically significant correlations were found

to exist between achievement, as indicated by the comprehensive exami­
nation grades, and (l) Cooperative Test of Reading Comprehension decile
rank; (?) ACE Quantitative score decile rank; (3 ) ACE Linguistic score
decile rank; (H) ACE Total score decile rank; and (5) quartile rank in
high Bchool graduation class.

A negative and statistically significant

correlation was obtained between achievement, as measured by the compre­
hensive examination grade, and the homework performance reported as the
•oercentage of assigned problems done by the students.

The coefficients

of correlation between quartile rank in high school graduation class
and (l) term grade; and (?) comprehensive examination grade were posi­
tive and statistically significant.
The statistical inferences which can be made from the analysis of
homework performance, by commuting correlations with variables previous­
ly reported as being important, are that statistically significant re­
lationships are:
signments in

(l) positive between

percentages of the homework as­

the textbook reported read and

school graduation class;

(?) negative between

quartile rank in high
percentage of the home­

work assignments in the lecture syllabus reported read and ACE Quanti­
tative score decile rank and the Cooperative Test of Reading Comprehen­
sion decile rank; (3 ) negative between percentage of homework problems

76

reported done, ACS Total score decile rank, and ACE Quantitative score
decile rank, and positive "between class of high school from which grad­
uated and ACE Total score decile rank; and (U) positive "between average
number of hours per week reported studied for the course and quartile
rank in high school graduation class, ACE Quantitative score decile
rank, ACE Linguistic score decile rank, ACE Total score decile rank, and
Cooperative T«st of Reading Comprehension decile rank.
The survey of literature and the data of thi9 study do not corrobo­
rate the assumed relation between homework performance and achievement
unless we interpret the correlations which are low but statistically
significant as indications of some degree of cause-effect relationships.
However, due recognition must be made of a multiplicity of variables
which may obscure the meaning of whatever correlative relationships are
found between two observed variables.

The absence of correlation coef­

ficients high enough to be considered corroborative evidence of a causeeffect relation between homework performance and achievement, for exam­
ple, does not disprove the existence of such a relationship.

The most

that can be stated in general terns from the data of the investigation
is that while no trend or apparent pattern of any degree of correlative
relationship between the homework performance investigated and achieve­
ment was found, a consistent positive trend and apparent pattern of re­
lationships between achievement and psychological and reading tests
were noticeable.

A derivation of causes for the lack of the assumed re­

lationships and for the occurrence of other relations cannot be made
from the d^ta of this study.

77

The in.tercorrela.tion. of -Ul3 between performance on the lecture
syllabus assignments and performance on the textbook assignments, as
shown in Table XVI would be expected since the teachers and students
generally believe the lecture syllabus to be the most im-oortant reading
source in the course-

The intercorrelations between reading and prob­

lem performance data indicate that the students' homework performance
Is consistent with the course structure, in which the library readings
and -problems receive much less emphasis than the other homework.
The negative intercorrelations in Table XVI between the amount of
time spent in study and the various other aspects of homework reported
done may reflect the typical emphasis, by both teachers and students,
on the idea that so many hours of study are to be done for so many hours
of credit in the course, and little concern is shown for what is actu­
ally accomplished during the study time.

Perhaps the most important in­

ference to be drawn from the table of intercorrelations is that even
though erroneous and fictitious data cannot be identified and eliminated
from the survey,

the students actually reported no relation between the

time spent doing the homework and the amount of homework accomnlished•
Negative intercorrelations suggest the possibility of fictitious report­
ing of data by the students.

I

7*

TABLE XVI
INTER CORRELATIONS OF HOMEWORK PERFORMANCE

Per cent
of text
reported
rend

Per cent
of lec­
ture
syllabus
reported
read

Per cent
of 11brary
readings
reported
read

Per cent
of problems
reported
done

Average
number
of hoars
Der week
reported
studied
for the
course

,125**

-.193**

Per cent of text
reported read
Per cent of lec­
ture syllabus
reported read

,2 6 6 ** '

Per cent of li­
brary readings
reported read
Per cent of
oro'olems re­
ported done

♦Statistically significant at .05 level.
♦♦Statistically significant at .01 level.
♦♦♦Statistically significant, at .001 level.

-.0 9 ?

-.011

- .0 6 1

79

C. HOMEWORK PERFORMANCE OF THE MEAN
STUDENT OF THE INVESTIGATION
The analysis of the data includes the statistical means "by which
the "behavior of the "typical student" may he inferred.

Even an assump­

tion that students exaggerate considerably in reporting the type of in­
formation solicited by the questionnaires of this sludy does not pre­
vent certain interpretations of the statistics derived from the data.
This is especially so when the means reveal that, even with the nresumed
operation of exaggeration and its unknown effects, the student is found
to be deficient in the nerformance of assignments when the expectations
of the instructors are used as the criteria.

Instructors, in the very

act of making up assignments and then distributing them to students are
giving overt expression to the expectation that
least read all the assignments.

r 11

students will at

Deficient performance on the homework

assignments is apparent from the data of the means and medians in Figure 1.
The performance of the student, during
classes during which the data of this study

the nine weeks of the

term of

wasreported by students and

up to but not including the week of the assignment of the term grades by
the instructors and the comprehensive examinations, appears to be con­
siderably less than is expected by the instructors making the assignments.
If such performance can be assumed to represent a reasonably accurate
account of what actually is done by students, an investigation of all
relevant factors seems to be a prerequisite
ments of the value of homework assignments.

ofthe forming of any

judg­

so

Figure 1
THE MEAN STUDENT OF THE INVESTIGATION

Characteristic

Statistical Mean

Percentage of homework assignments in the
textbook reported read

5^.51

Percentage of homework assignments in the
lecture syllabus reported read

6 7 .OS

Percentage of assigned homework problems
reported done

31.67

Percentage of assigned library readings
reported read

U?.g5

Average number of hours per week reported
studied for the course
Class of high school from which graduated

2.89
5 (median)

Quartile rank in high school graduation
class (from top)

1.77 (median— 2)

Cooperative Test of Reading Comprehension
decile rank

5 .U0

American Council Psychological Quantitative
score decile rank

5.50

American Council Psychological Linguistic
score decile rank

5.3*+

American Council Psychological Total score
decile rank

5-37

Term grade

1.99 (slightly
under C)

Comprehensive examination grade

2.22 (about
C+ )

gl

D.

SCHOLASTIC GROWTH OP THE POPULATION

Even though the students* uerfornance of homework assignments has
no apparent relation to the grade earned on tests over the assignments,
can the inference also he made that no learning occurred as a result of
experience in the course?

Evidence is available from columns 2 and 3

Table X7II from which the amount of learning or the inference of scholas­
tic growth for the respondents of this stuhy can be made directly.
In the spring term of 1950* contrary to usual practice, the compre­
hensive examination items were so arranged that the two halves. Part I
and Part II, were alike in the subject matter covered and in the types
of knowledge and ability required to answer the questions*

Thus each

half of the examination was a fairly representative sample of the entire
examination.

The change in procedure from the usual practice was done

so that half of the examination (Part I) could be used the following
fall as a pre-test or olacement test.
The grades of the students writing the examination in the spring
were, of course, based on their scores for the whole examination since
they had completed the three terms of the course.

The scores of these

third term students on Part I of the examination were also equated with
the grade distribution for the entire examination to serve as a basis
for comparison of future scores when Part I was used as a pre-test, as
in the fall term of I9 5 O.
Table XVII shows the grade equivalents of the scores (1) made by
a group of students who had completed three terms of the course and who
wrote the examination which was arranged in the two similar halves as

82

TABLE XVII
GRADE EQUIVALENTS OF SCORES MADE BY STUDENTS ON A BIOLOGICAL SCIENCE
PRE-TEST AND ON COMPREHENSIVE EXAMINATIONS*

Grade

Third-Term Students
Spring 1950
*

Entering Freshmen
Fall 1950
£

Third—Term Student
Soring 195^
*

A

5-1

0.0

5.1

B

26.7

0 .1

25.9

C

U9 .I

u.u

U9 . 8

D

15-8

15.5

1 5 .6

F

3.**

80.0

3.6

*Note the similarity of the distributions of the grades for the
comprehensive examinations, spring 1 9 5 0
spring 1 9 5 1 »

explained above; (2) made by entering freehmen the following fall on
the pre-test made up from Part I of the spring comprehensive examination
as explained above; and (3) made by the same students who had taken the
pre-test and then took a regular comprehensive examination in the follow-

6
ing spring after completing three terms of the course*
Although the grade equivalents in Table XVII give a comparison of
the performance of different groups ot students on the same test in
columns 1 and 2, the examiners believe that the entrance examination
scores of the students show that they are comparable groups.

The data

show, for example, that only U.U per cent of the entering freshmen made
scores as high as those of students who received C on the comprehensive
examination.

If the assumption is valid that the two groups of students

in columns 1 and ? are comparable* then the scholastic progress asso­
ciated with three terms of coursework becomes apparent.
Data for the same students on different tests, pre-test for enter­
ing freshmen, fall 1950* believed by the examiners to be comparable to
the regular conro rehen sive these freshmen took in the spring of 1 9 5 ^
after completing three terms of the course, are reported in columns 2
and 3 of Table XVII.

If the assumption that the pre-test and the spring

1 9 5 1 comprehensive are comparable is correct, then the inference is that
students can perform much better on tests over the coursework after com—
oleting three terms of coursework than they can before doing the coursewcrk.

But the part or parts of the coursework which contribute the most

cannot be ascertained from such data.
~^Data for Tables XVII and XVIII were obtained from the Board of
Examiners, Michigan Stata College, East Lansing, Michigan.

Table XVIII contains data from the pre-test scores of entering
freshmen and scores made by a different but comparable group on the same
test after completing the three terms of coursework.

The medians and

means of the two groups, one taking the test before taking the course
and the other group taking the same test after completing the three
terms of work show a marked difference in performance*

Such differ­

ences sunoort the inference that students learn something during the
three terms of coursework and that this learning represents scholastic
achievement.
accounts
The
what the

Here again, the data do not indicate which course activity

for most of the change.
conclusions are that while no relation anoarently exists between
respondents did on homework assignments in the course and

sequent achievement,

they made definite progress in subject matter

scholastic achievement which was probably due in large part to the
classroom experiences during the three terms of coursework.

sub­
or

85

TABLE 171II
MEASURES OF CENTRAL TENDENCY AND DISPERSION OF PRE-TEST SCORES OF EN­
TERING FRESHMEN AND S 00RES OF A GROUP OF STUDENTS
ON THE SAME TEST AFTER COMPLETING THE COURSE

Measure

Entering Freshmen
Fall 1 9 5 0

Third-Term Students
Spring 1950

0 -1 1 6

Hg-129

Mean

57.8

S8.7

Median

58.3

95.9

Standard deviation

17.3

1 2 .g

N

2?30

2063

Range

86

CHAPTER V
SUMMARY OP FINDINGS, EDUCATIONAL IMPLICATIONS, AND LIMITATIONS OF THE
FINDINGS, WITH SUGGESTIONS FOR FURTHER STUDY
A, SUMMARY OF FINDINGS
Homework assignments In the Investigation. During this investi­
gation of the relation between the amount of performance of homework
assignments and subsequent achievement, the prevalence of homework as­
signments in schools and colleges was noted as well as the inherent
assumption of a cause-effect relationship between the amount of home­
work assignments done and the grade or mark achieved.

This practice

and assumption Beems to have been undisputed and uninvestigated at the
college level, except for a few studies of hours of study time and de­
grees of achievement which produced contradictory findings.
The survey of fifty educational methods tcx.itcoks, which was
done by the writer, indicated that the authors of the texts, as well
as the teacher training institutions which use the texts, continue to
include homework assignment as a suggested teaching method in spite
of the weight of the research findings from elementary and secondary
school investigations which have discovered the value of homework to
be questionable, at best.
The homework assignment under investigation is exemplified by the
classroom situation in w>ich all students are given the identical as­
signment by the instructor, and, after what is supposed to be an adequate

87

time interval for performance of the assignment, are given an identical
test over the assignment by the instructor, from which grades or marks
are assigned.
Homework performance in the investigation.

The investigation of

the relation "between the amount of performance on homework assignments,
as reported "by students through a log—questionnaire survey method, and
subsequent achievement, used data pertaining to students in a college
science course.

The data from the survey do not corroborate the assumed

positive correlative relationship between the amount of homework done
and achievement.
An anlyBis of homework performance based on such qualitative var­
iables as psychological examination deciles, reading ability deciles,
quartile rank in high school graduation class, and class of high school
from which graduated, revealed that the extent to which such variables
indicate quantitative and linguistic ability has no anparent relation
to the amount of homework performance.

Some relationship was indicated

by a low positive correlation between the percentage of homework prob­
lems renorted done and the size of the hiffr school from which graduated.
A high positive correlation between quartile rank in high school gradua­
tion class and the number of hours per week studied for the college
science course may indicate a consistency in the pattern of study time
for high school and college.
The typical student in the investigation, as interpreted from
mean performance, was found to be deficient in performance of the homework

88

assignments when compared with the expectations of the instructors.
Mean performance varied from a low of 32 P er cent of the assigned
problems reported done to the highest mean of 67 per cent of the lec­
ture syllabus assignments reported done.

The mean of the reuorted

average time spent for study of the course was 2.89 hours per week for
the three— credit (per term of classes) course requiring two hours of
lecture and two hours of laboratory classes each week.
Relation of the amount of homework reported done and achievement.
No correlations large enough to have predictive value were found between
achievement in the college science course observed and the reported
amount of the homework assignments done by students.

Nor did the amount

of time studied for the course have any predictive value for achievement
in the course.
The data of the investigation corroborate the findings of reported
studies of the relation of homework to achievement which were done at
the elementary and secondary school levels.

The findings of this in­

vestigation at the college level may be added to the weight of evidence
which does not support the assumed relation between work on assignments
and achievement.

The exceptions were a few studies at institutions of

higher learning which produced contradictory findings.

The summation

of the reported findings, together with the findings of this study, does
not corroborate to any extent the cause-effect relationship heretofore
assumed to exist between the amount of homework done and achievement.

89

Analysis of achievement.

The literature and data of this study

support the contention that the amount of work done on homework assign­
ments cannot "be used for prediction of achievement in the course.
Achievement, as shown by the data of this investigation, appears
to have a higher relation to reading ability first, general college
ability second, and, to a less extent, library reading assignments re­
ported done, than with the other variables of homework performance
investigated.
The data of this study corroborate to some extent the findings of
Seder that:
Correlation between ^-scores and achievement test scores in
the sciences are lower than correlations between L— scores
and science.
Although sciences are ordinarily regarded as
closely related to mathematics in high school courses they
are quite verbal in nature. Perhaps the scores of students
in scientific curricula in colleges might show more corre­
spondence with Qyscores than with L-scores.*Although Seder's findings were based on pupil performance in high school
grades nine through twelve, the data of this investigation of homework
perforr.ar.ee show the L scores to have a higher relation to achievement
in the college science course observed than do the Q scores.

The data

indicate that the suggestion which Seder makes concerning college science
curricula is not applicable to the science course observed in this study
The findings of this investigation of homework performance support
the conclusions of Segel and Oerberich:

Seder, "The Reliability and Validity of the ACE Psychological
Exairi,'' J o u m a l of Ednca ti onal Research, 3^:100, Oc tober, l^ho.

90

It may "be concluded
logical Examination)
prediction purposes
since its cower for

that
Lest
when
such

the American Council (Psycho­
should not "be UBed for differential
college marks are the criteria,
use is negligible.^

The correlations between achievement of students in this investigation
and the ACE decile ranks varied from a low of .092 to the highest ob­
tained which was .^13*

None were high enough to warrant prediction of

achievement from the ACE deciles on an individual basis.

Counseling of students about homework.

The literature and the data

of this study do not support a counseling policy of recommending that a
student found deficient in scholastic achievement should devote more
time to study outside the classroom.

The conclusion most nearly con­

sistent with the literature and findings of this study is that the
quantitative aspects of homework performance,

i.e., the amount of time

spent in study and the amount of the assignments done, are probably
of minor importance to achievement in comparison with the quality of
the homework performance, which remains to be investigated.
The low correlations obtained in this study correspond with the
conclusions published by MncFhail:
Inferences made in the manual for the ACE Psychological Ex­
amination for College Freshmen pertaining to the use of the
Q, and L scores for counseling and sectioning purposes can
not be safely assumed to be applicable to the situation in
a particular institution and,... any given institution would
do well to discover the local pertinence of these scores be­
fore putting them to any such use.3
^D. Segel and J* R. G-erberieh, "Differential College Achievement
Predicted by the American Council Psychological Examination," Journal
of Applied Psychology, 17*6U5, December, 1933^A. H. MacPhail, "Q and L Scores on the ACE Psychological Examina­
tion," School and Society,
September, 19^2.

91

Scholastic growth and homework performance.

There is evidence

that the students In the Investigation demonstrated learning and scho­
lastic growth during the time spent in the course, although the evi­
dence does not show what aspect of the learning situation was the most
important contribution to such growth.

But apparently the scholastic

growth has no relation to the amount of homework done as analyzed in
tniB study.

Since some degree of scholastic growth is apparent, the

inference may "be made that classroom experiences, together with the
variable effects of quantitative, linguistic, and reading ability, are
more closely related to achievement than the amount of homework done.
B.

EDUCATIONAL IMPLICATIONS OF THE INVESTIGATION

Many teachers are familiar with the explanation of low scholarship
by students who sa,v, H I just can't understand why my grades are so low.
I try so hard and study all the time; but it doesn't seem to do any
?ocd."

Such statements are frequently received with much skepticism

by teachers and faculty counselors, not because the student is delib­
erately dishonest, but because the teacher believes the student can­
not make a valid estimate as to what amount and kind of studying is
actually done.

Many teachers have noticed that students tend to over­

estimate the number of hours studied and to underestimate the hours they
pive to extra-curricular activities and leisure.
Some of the facility are often operating under a much different
opinion.

Faculty members may attribute scholastic difficulties to

either lagrardliness or lack of intelligence.

Some staff members

92

frequently feel that students actually attend classes very few hours
during the week and "believe that students can always be found in nearby
eating places, places of recreation, public lounges on or near the cainmis, and
The

very seldom in the library or at a study desk.
findings of this study, which show the students to be consid­

erably deficient in performance of homework when compared with the ex­
pectations of the teachers, may have certain implications for staff
members.Staff members tend to believe that achievement
the amount of homework done, but permit the

is related to

typical student with de­

ficient homework performance to achieve a passing grade.

There seems

to be an inconsistency in a practice which involves the making of home­
work assignments, the making of tests over the homework assignments, and
then the subsequent passing of students having deficient homework per­
formance, as shown by this study.
The weight of the evidence from the literature and the findings
of this study indicate that poor achievement cannot be assumed to re­
sult from lag-ardliness or lack of intelligence.

Teachers, would do

well to re-examine the assumption that homework assignments have a nlace
in educational methodology.

Perhaps one idea that is reinforced by the

entire investigation of the relation between work on homework and sub­
sequent

achievement is that no teaching or administrative method or

oolicy should be exempted from continuous evaluation in terms of its
ourported objectives or results.
The findings of the study seem to show that students achieve scho­
lastic growth in the course observed even though the relation of the

93

grade received with the amount of homework done ia negligible.

One

implication, which may he of much value, ia that the classroom experiences
may he more closely related to scholastic growth and achievement than
any of the other variables analyzed.

C. LIMITATIONS OF THE FINDINGS WITH SUGGESTIONS F O B FURTHER STUDYDetailed elaboration of statistical refinement has not been employ­
ed to demonstrate the scientific reliability of each step in this study
of the assumed correlative relation between work done o n assingments and
achievement.

The design of the study was derived from precedent reported

in the literature for using ouestionnaire data and simple correlations in
the Investigation of aspects of student performance on assignments in re­
lation to achievement.

Competent statistical authority was consulted

which suooorted the assumption that for purposes of thi s initial inves­
tigation at the college level the simple correlations would orovide
evidence from which it could be ascertained whether or not a more re­
fined analysis was desirable in future studies.
While the log-questionnaire survey method developed and utilized
for gathering data of this study may have elicited more or less accurate
renorts from the students, the absolute validity of the method of sur­
vey cannot be demonstrated.

The reliability with which the students

could cony the data from their study logs to the questionnaires was found
to be high.
ascertained.

The accuracy of the reports of the students cannot as yet be
However, the recommendation should be made that a future

9U

study could investigate the validity of the reports with the log—ques­
tionnaire method by giving some examination over some part of the as­
signments made and reported as done by students to ascertain whether
the test performance was consistent with the reported homework perform­
ance.
The study log was constructed with a column in which the weekly
blocks of homework assignments .v'r.i to he written.

Students wore di­

rected to place a mark next to trie block of assignments on the study
log as the assignments were comoleted and to make a second mark to
indicate that the comoletion of the assignment had been renorted on the
weekly questionnaire.

Since the weekly questionnaires revealed that

students indicated occasionally that they had spent time studying but
reported no assignments comoleted, there is a oossibilitv that some
part of the assignment may ’
nave been done; but since the assignment
was not actually com 'leted, the part that was done might never have
been reported.

The extent to which the possibility of not reoorting,

when onrt but not all of the assignment was done, has affected trie cor­
relations is unknown.

The negative intercorrelations of Table XVI may

be partly the result of this loophole in reporting.

Future studies in

which a stud;/ log is to be used as a record of the amount of the assign­
ments done should provide for the students recording of the number of
pages done rather than require them to report blocks of pages done after
completion of the assigned block of pages, or require then to do both.

95

The apparent seriousness with which the respondents took the inves­
tigation, and the degree of dependability of the data obtained, must
rest to a lerge degree upon the cumulative effects of careful planning
and development of the survey method and the subsequent solicitation of
information.
hibited

bs

However, such mute testimony cannot be distilled and ex­

evidence of the validity of the data.

The findings of this

3tudy are based not only upon the experience of the pilot studies and
the careful development of the log-questionnaire method but also upon
the more tangible evidence that the questionnaire data is an accurate
reproduction from the study log.
Although the problem of investigating the occurrence of correla­
tive relationshios between achievement and homework performance has been
initiated here, the corollary problem of explaining why such relation­
ships were or were not found remains to be studied.

The contention may

be raised that the low correlations between the amount of homework done
end achievement are due to mutually exclusive variables.

For example,

a student with a high ACE decile rank would be expected to achieve a
high grade regardless of the amount of homework done, according to such
a contention.

Examination of the distribution of reported homework of

all res undents in the ACE tenth decile (high) and all of those in the
ACE first decile (low), as shown in Tables XIX and XX, suggest that when
the intelligence variable is held constant a very low correlation might
be found between achievement and performance of homework.

If the con­

tention of mutually exclusive variables were correct, then the achieve­
ment of the ACE tenth decile students would always be high and the

96

ac hievement of the ACE low decile

students would always he low.

sample of 17^ of the U 73 respondents,
deciles and gr ades with
in Table XXI.

A random

used for the di s tri but io n of A C E

the am ount of study time held

constant,

is shown

An inference which may he made from the di st ri but ion is

that a small correlation

exists betwe en A C E deciles and achiev em ent when

study time is held constant.

Au-ain,

clusive va ri ab le s were correct,
••rid ach ie vem ent

if

the con tention of mu t u a l l y ex­

then t.ue co rrelation b e tw een ACE deciles

would be ext re me ly hi,~h.

The st u d y of homework p er fo rma nc e miyht produ ce more val uable re­
sults if it wore extended
of m a k i n g assignments,

to investigate

the effects of different meth ods

of different types of a s r i y n m e n t 3 , of the reasons

why students do or don't do homework,

and of

the social implications of

coercive h o m e w o r k assignments.
If the per for ma nc e of
co-op-'red with
ve st iga ti on

the students

is fmind to be deficient when

the exp ectations of the teachers or administrators,

is indicated

to d i sc ov er what varia ble s are ^ m d u c i n y

an in­
such

~ sit ua *ion.
Fu tu re
ac hi ev e me nt

studies of re lationships be tw ee n homework
could subject

work as oif-nmen t s and

exper ime nt al

sections

performance and

to v a ri o us kinds of ho me ­

the subsequent h o m ew or k p e r f or ma nc e and ac hievement

analyzed to determ in e the extent of

the re lation

h o m e w o r k performance

to achievement.

1

97

T A B L E X IX
D I S T R I B U T I O N O F TERM GRADES A N D H O M E W O R K P E R F O R M A N C E F O R THE
R E 3 P 0'HDEHTS IN T H E A M E R I C A N C O U NC IL P S Y C H O L O G I C A L
E X A V I N A T I O N T E N T H D E C I L E GROUP

C|} H

ere As-’ipnrade m en tv.

A

Textbook
Syllabus
Lib ra 17/
P robl em e

_________________
E o m e ^ ork perfcr.Mflr.ce In te r v a l s_____
0— 1
done 7 0 — 39/k done Uo— 5 9 ^ done
60-7 9 % done
SO— 100?

1
1
1
0- 1.9 hrs

T:

li-R.q lifS

1

Te xt bo ok
SylInbus
Library
Pr o b l e m s

1
1
1

T

h - “ . G hrs

*- c . 9 'r.i'9

Tex tbook
Syllsb-;p
LJ Hr a r y
P r ob le ms
Hrs./wte.

r

1

1

r-3.9 hr?

Hrs. /wk_._ ____ _____________ ?____

Hr r.. /ttic .

\-

1

1

Tertbr ok
Syllnbup.
Lib ra 1*7,'
Pr oblems

1

L
H
O - l . o hrs
P
7

*7

l
9- 7.9 • •0

H r s . /wk.

L

cr
T

i
1

a.

~* "h • ^ r.7*5*
3

?• 9 hr«
n

1

’ 7 . R hr*
7

"4 r.rc
.

..A.

1

1
:>-t. 9

t

rr

L^.r
r.
-m _•

]~lT * e

v rr

H r s . /wk.

F

3* hrs

1
7

1
7
0— 1 . G > i's

Te xt boo k
Syl k b u o
Library
P r ob le ns

c
3

1
1

•>

1

9— 1.9 h r p
.
. 3

34 hre

1

>’
iA.

hL.r 0 viy*

34 hre

2
1
G
A

1
1
'— •1.9 hre

2- 3.9

► _ ▼* C;

U— 3.. 9 h rs

'>-7.9 hrs
h

'<4* hrs

9H

TABLE XX
S 7 R I B U T I 0 N O F T E R M GRADES A N D H O M E W O R K PEEFCK1.1ANCE F O R THE
30 RESPONDENT'S I N T H E A M E R I C A N C OU NCI L F S Y C H O L O G I C A L
E X A M I N AT IO N FIR ST DE C I L E GROUP

'
O H

era A s s i g r:—
rade m en ts __________ Q— 1 9 ^ done

A

Hom e wo rk p er fo rma nce Intervalo___________
? Q ~ 5 9 ^ done uQ— *39^ do ne
£0— 795^1 o n e
30— 1005?

T e xt bo ok
Sy ll ab ua
L i br ar y

Problems
o— l •

3«9 hre

L— E.9 hrs

<-■—

7 ,cj hrs

3^. hrs

H r e . /wk.
Text > 00k
S y l 1 fll’is
Li brary

S

Problems

1
0-1

C

'j
3

i.

1

?

.

H r s . /wk.

1

Textbook
Syl labus
L ib rar y
problems

1

Textbook
S y ll ab us
Li br ar y
Problems

I
7
IO
9

!fe

h~~.G Hrs

3^ hrs

l’
7

3

P

I

^
*4
7
0— 1.9 hrs
P— 3*9 hre
______ H r s . /w k .__________ 6 _____________

7

‘'— 7.9 hrs

7
1
7

0
3
1
7
p
I4
L
1
U— 5.9 hrs
6-7.9 hrs
3* hre
1 __ __________ ____________________
-4

L

7

5
10

1

1
3

L
6
1
7
0 — 1.9 hrc
'’-7. 7 hrs
H.9 hrs
6- 7.9 hrs
3+ hrs
______Hrs. /w k . _________3_____________ ^_______________________________________________
Textbook
Syl Ia bug
Li b rar y
Pro b le ms
J •

Hrs./wk.

7

r- rs

7 —^ . 7

k r «

**—7 .

1

hrs

t —7 .

7

hrs

3+

h rs

99

TABLE XXI
DISTRIBUTION OF TERM GRADES AND AMERICAN COUNCIL PSYCHOLOGICAL EXAMINA­
TION DECILES OF A RANDOM SAMPLE OF 17* OF THE RESPONDENTS
ACCORDING TO THE NUMBER OF HOURS PER WEEK OF STUDY

Term
Hrs
per
grnde
i»celc
stud­
1
ied

2

3

u

5

6

1
1
3

1
1
5
2

1
1
1
8

1
6
4
3

2
3
2
1

1
3
1
1

2

3
1

1
2
2
2

2
1
1

▲
0-1.9

B
C
D
F

?-3.9

A
B
C
D
F

^5-9

6-7.9

S+

A
B
C
D
F
A
B
C
D
F
A
B
C
D
F

ACE Deciles

1
3
3
3

7

2
1
1
1
2
3
5
2

8

9

1
3
5
1

3
3
4

2

1
1
1

1
3

1

1

1

1
2
1

10

3
2
1

1
1

1

l

1
1

1
3

1
1

2

3
1
1
1

1

U

2
1

1
1

1
3

2
1

1

1

I

loo

Two major qu es ti on s ar i s e from the n a t u r e of this inv es t ig at i on and
t’-e l i m i t a t i o n s of the findings.

Are

the low c or re la ti on s found b e t w e e n

the a m o ’int of h o m e w o r k do ne and a ch i e v e m e n t
a valid an a l y s i s of
are the data

although

the h o m e w o r k p e r f o r m a n c e — ac h i e v e m e n t

erroneous?

tion that val id

in the co ll eg e science co urse

The

relotion,

i nv e s t i g a t i o n was designed with

the a ssump­

dat a w o u l d he o b ta in ed b y the loy— cues tionn.a ire m e t h o d

the de y r o e of v a l i d i t y has not been asc ertained.

is assu me d

or

If the data

to be re a so na bl y v a l i d then man;.' problems for future study

can be d e ri ve d

from

the r e s p on de nt s was

this study.
found

tween a c h i e v e m e n t and

A m p l e e vi de n ce of sc h o l a s t i c growth of

in suite of the a m n r e n t

the a m o u n t of h o m e w o r k

The se cond qu est io n w h i c h arise s

lack of relation b e ­

done.

from the study is,

whnt is

the

na ture and ex ten t of the c o n t r i b u t i o n of classr oo m ercoeriences to the
a ch ie vem ent

or scho las ti c yr ow th of students?

The f i n d i n g s of thi3 study do not di s r r o v e
relative re la t i o n s h i p b e t w e e n p e r f o r m a n c e

the e x i s t e n c e of a cor­

on h om e w o r k as si gn me nt s and

■p '.oeouent r>er f o nnance on test? over t- e as si m i m e n t s . but n e i t h e r do
the da ta c o r r o b o r a t e such an assumed

re la t io n for the gr o u p investigated.

A n obvious

to all

recommendation

to be m a d e

o J~ a s s i g n i n g h o m e w o r k is that nearly all

of

teachers who m a k e a p r a c t i c e
the report ed

i nv es t iga tio ns

cast c o n s i d e r a b l e doubt on any a s s u mp ti o n of a co r re la ti ve rel at i on sh ip
b et we en h o m e w o r k done and

achievement.

sig nment of ho m e w o r k mi yht do well
.:ie th.o d of assignm en t,
classroom

lar tea c h i n ” rr»«t}od.

to examine

a ’"d the effects

si tua ti on where

Ev e r y

te.ic'mr p r a c t ic in g the as­
the type of a s s i g n m e n t , the

of the as si gn me nt

the h o m e w o r k a ss i g n m e n t

in the p a r t i c u l a r

is em pl oy ed as a regu­

101

APPENDIX
I.

II.
III.

Table

Lo^?-Questionnaire Materials
Clerical and Tabulating Materials

i

102

APPENDIX I
TABLE

NUMBERS OF PAGES IN EDUCATIONAL METHODS TEXTBOOKS AMD NUMBERS OF PAGES
INDICATED BY THE INDEXES AS CONTAINING DISCUSSIONS OF ASSIGNMENTS

Educational methods textbooks, authors,
and sources

Total
number
of
pages

Number of
pages indexed
as discussing
assignments

/imack, John C. and A. R. Lang, The
Beginning Teacher (Chicago: Houghton
Mifflin Co., 19??)

U7S

6

Avent. Joseph E. , Excellence and Errors
in Teaching Methods (Knoxville: Joseph
E. Avent, 1931)-

569

31

Avent. Joseph E . , Beginning Teaching
(Knoxville: University of Tennessee,
19^7).

596

lU

Begley, William C. and John A. Keith,
An Introduction to Teaching (New York:
The Macmillan Company, 1977).

Uoo

p

Bagle.v, William C. and Marion E. Mac
Donald, Standard Practices in Teaching
(New York: The Macmillan Company, 193*)*

179

11

Blackhurst. J. Herbert, Directed Ob­
servation and Supervised Teaching~TBoston: Ginn and Co., 1925)*

U?0

9

Bossing. Nelson, L . , Progressive Methods
in Teaching in Secondary Schools (Boston:
Houghton Mifflin, l'tho).

7?q

39

Brink. William G. , Directing Study Ac­
tivities (New York: Doubleclay, Doran and
Co., Inc., 1977).

73*

U6

103

APPENDIX I, TABLE, Cont'd.

Total
number
of

Source

P.a^ el.

Brubacher, A. R., Tefesslona and Practice
Century Co., 1927).

Number of
pages Indexed
as discussing
assignment s

701

Burton, W. H. , Supervision and the
lairoreaest of Teaching (New York:
D. Apuleton Century Co., 1922).

RIO

11

Burton, W. H., The C-uldance of Learn­
ing Activities (New York: D. Aopleton
Century Co., lPLIi).
"

601

13

Butler, Frank A. , Tr.e I:rmroverrent of
Teaching in tine Secondary Schools
(Chicago: University of Chicago Press).

3~?

2?

Carr, W. 3-., and J. Waa'e, The Lesson
Ass lgnnent ( Stan ford Ur*iversity, Cali fornia: Stanford University Press,la31'*

9?

Colvin, Stephen, Introduct Ion to High
School Teach ir.g (New York: Th e Macmillan
Co.”, i-TTTT

U31

7

Crawford, Claude C. , Hew To Teach (Los
Angeles: Southern California School
Book Deoositry, I9 3 S).

Rll

10

Dearborn, Ned K., An Introduction to
Teaching (New York: D. Appleton Century'
Co.', 192:).
*
Douglass,

K. R . , Mo de rn M e t h o d s

237

1

9 LL

i?

in H 1gh

School T e a c King (B o 3 1 on : H o u ~ h t o n Mif­
flin Co.. Y 920T.

Fontaine, E

Jla-ke, tVavs to Better
Secon da rv School (BosIf— c--log in
ton: Dir and Co., 19 v*-

'71

10U

APPENDIX I, TABLE, Cont’d.

Source

Total
nutiber
of
papes

Number of
pages indexed
as discussing

assignments

Ford, Frederic'-: A. , The Instruc­
tional Program (New York: Frentice—

Hall, Inc.7 1-33).

?71

pU

Foster, Herbert H . , Principles of
Teaching in Secondary Education (New
York: Charles Scribner’s Sons, 1321).

3^7

6

GiLmore, Marquis E. , Exemplifying
Good Clas?room Methods and Proce­
dures (Boston: Trie Christopher pub­
lishing House, 19^1)•

P3?

1

Goettir.g, M. L . , Teaching in the Seconrary School (New York: Prentice-Hall,
Inc.ri9U?).

?1?

3

Hsll-^uest, A . L. , Supervised S t u d v ,
(Hew York: The Macmillan Comcany,
191S).

^73

3?

Keer, Amos L . , Steps to Better
Teach ins' ('Tew York: W. W. Norton and
Co"., 1°37 ) .

31c

17

Holley, Charles E. , High School Teach­
ers * Methods (Champaign, 111: The
Garrard Press, 1937)*

17

Knudsen, Charles W. , Evaluat1on and
Improvement of Te^c' lng (New York
Doubleday, Doran and Co.,Inc. ,193‘?)•

53^

u

Lancelot, W. E . , Permanent Learning
(New York: John Wiley and Sons, Inc.,
inhh).

??1

K

Maxwell, C. H . , and W. C. Reusser,
Observation and Directed Study (New
York: Prentlce-Hall, Inc., 1^79)•

hyu

11

105

APPENDIX I, TABLE, Cont»d.

Source

Miller, Harry L. , Directing Stady
fNe-.T York: Charles Scribner's Sons,
Inc... 19?^).

Total
number
of
pages

377

Number of
pages indexed
as discussing
assignments

Ul

Monroe, W. S. Directing Learning
in the High School (New York: Doubleday, Page, and Co., 19??).

577

7

Morrison, Henry C., The Practice of
Te -ching in the Secondary School (Chi—
cago: University of Chicago P r es s,19?6)„

?6l

0

Mursell, James L. , Successful Teaching
(New York: McGraw-Hill Book Co., Inc.,
19^6).

33?

U

Nutt, Hubert W. , T:.e Supervision of Ins t.rue 11on (Boston: Houghton Mifflin Co.,
1??0).

?77

Parker, Samuel C. • Method3 of Te-1chlng
(Boston: Ginn and Co., 19?o7T

5?9

Reagan, G. W . , Fundamentals of Teachir.^ ( M e w York: Scott, Foresman and Co.,
1?"*?).

lU

15

Reeder, Edwin H . , Simplifying Teach­
ing (New York: Laidlaw Brothers, 19?^).

19?

11

Reeves, Charles E . , Standards for High
School Teaching ( N e w York: D. Apuleton
Century Co., 1 ')'*h .

‘
;l h

*»

Rink 1 Thomas M . , Principles and Prac­
tices of Teaching in Secondary Set'cols
(New Y o r k : American Book Co., 19^lT*

7??

16

Rueciger, William C. , Teaching Procedures (Eogton: Hougiton Mifflin Co.,
1939).

Ujp

?fc

106

APFENDIX I, TABLE, Cont'd.

Source

Total
number
of
pa^es

Number of
pa/res indexed
as discussir£
assifnmer.ts

Russell, Charles, T each inf: for To­
morrow (New York: Prentice-Hal 1 , Inc.,
T^ffTT

u? 7

11

Stewart, W. F. , Methods of G-ood
Teachir k (Colunbua : The Chio State
University Press, I90C ) .

??C

?

Stormtard, Martin, Progressive Methods
of Teaching (Eoston: Hous'd'.ton, Mifflin
coi, r V ? ) /

331

11

Thayer, V. T. , The Passing of the Reci­
tation (Boston: D. C. Heath and Co.,
i.

?oo

lU

Thomas, F. W. , Prl rcloles and Techr.igues of Teaching CChicayo: Houghton
Mifflin Co., 1'

S

apples, Do’.i^lng, Procedures in Hi f±i
S c h c o 1_ T e a c ' - . i n /'

( N e w Y o r k : m: e Maerr. i 1 1 a n

Co '.toan;', 1 ^ 7 ;.

}.~b

1 -:

Wilson, H. E . , and 1. M. W:Ison, The
Motiva tion o f School Work (New Y o r e
rk.'^r.tcn Mifflin Co., 19? 1).

?73

n

Wilson, H. B. , 3-. C. Hyte, and. H. 3.
L".ll , M o d e m Methods in Teach ing (New
York: Silver Bv.rd e 11 and Co., 19?^/ •

P'-iC

?1

Yirir.kle, Wi Ilian'. L. , and Winfield D.
A m entrant, Di rected Observation and
Tenchinft lr. Secondary School's Thew York:
Tie Macmillan Co., 197?T^

39'.’

?u

I

107

APPENDIX, TABLE, Cont'd.

Total
number
of
nages

Source

Number of
pages indexed
as discussing
assignments

Wynne, John P ., The Te-cher and the
Curriculum (New York: Prentiee-HalT,
Inc., 1937)•

UUO

3L

Yoakum, Ceral A., The Improvement o f
the Assignment (New York: The Macmil­
lan Coimanv, 193?).

?9&

?9S

21U33

1079

TOTALS

10 8

APPENDIX II
LOO-QUESTIONNAIRE MATERIALS
BIOLOGICAL SCIENCE STUDENT'S TERM PROGRESS SHEET AND STUDY LOG
(1) Data for the Investigation of the relation "between nerformance of
outside assignments and achievement must be solicited from students.
You are asked to contribute to the study of this problem by reporting
on w e e k l y questionnaires the amounts of the various assignments which
you do. Only through your cooperation is such an investigation nosr,i ble.
(T) The information you renort on the weekly questionnaire in the lab­
oratory class will not be used in any way whatsoever to determine or
influence your grade in this course or in any other course.
(3) The value of your contribution to the investigation will be deter­
mined largely by the accuracy of your reporting on the questionnaires;
since the researcher is the only one who will comvile the data and his
primary interest is in the entire groun results, your cooperation will
be much apnreciated.
(U) The investigation involves an appraisal of the homework assign­
ments made by instructors, the work done on the assignments by students,
and achievement of the students as measured by grades or marks.

109

APPENDIX II, LOG-QUESTIONNAIRE MATERIALS, Cont'd.

STUDY LOG*

Assign­
ed
pages

Lee. Syllabus
Libr.Rdng.
I
Assign- R| Hi
p
ed
e p
i
pages
a r

Stud.Probs
AssignH D
ed o
probe. n

Hrs.Studi ed
No.of hrs

Apr

lU

21
28
•May "be used as a study aid when the assignments are recorded to
done; "by noting the completion of the assignment at the time studied
the re-nortinfr of the data will "be more accurate.

"be

I

110

APPENDIX II, Cont *d.

qUESTICNUAlRE SCHEDULE
Week
of

Approx.
time

Apr.
?

10— 30
min,

Establish rapport concerning investigation; distribute
and dipcr.cs term progress sheet and study log.

9

5 min.

Die tribute and collect 1st weekly questionnaire.

16

5 min.

Distribute and collect Pnd weekly questionnaire.

~*3

5 min.

30

10 min

Procedure

L ' .t r V:-a-..'•
•

: u; c l.U o t,

r . . •(. k L y

u«. -;t I

ire .

1. Distribute ogid collect Uth weekly questionnaire.
Distribute arid collect written "Re-take” question­
naire printed in green; use last 5 min. of class
-oeriod.

May
7

5 min.

Distribute and collect qtb weekly questionnaire

lU

5 min.

Di stribute and collect 5th weekly quo s t ionnaire

pi

5 min.

Distribute and collect 7th weekly questionnaire

O
r?

5 min.

Distribute ar»d collect 8 th weekly ques t ionnaire

Exulana t ior.y :
A. The regular weekly reporting on the questionnaires will require lit­
tle c Is s b time if forms are at students* seats when they arrive;
students may be encouraged to take the oues tionnai re blanks with them
to fill in during the week as they study.
Use of the questionnaire
for taking of roll is suggested to save time and to assure regular
reporting of data.
B. The "Re— take* renort (5th week) must be carefully handled and suffi­
cient time should be allocated to permit accurate reporting. The
re-take consists of asking the students for the same information they
have already reported at the first of the hour.
The re-take data
will be used to determine an estimate of reliabi1ity.

Ill

APPENDIX II, QUESTIONNAIRE SCHEDULE, Cont'd.

C, Whenever you feel that interest in reporting is lagging seriously,
you may take as much time as you feel necessary to encourage class
discussion of the investigation and what the students think the re­
sults will "be.
Your cooperation is very much appreciated as will he any criticism or
suggestion you may wish to make as the survey progresses.
Thank you*

11?

AK-ENDIX II, Cont'd.

BIOLOGICAL SCIENCE 1?? STUDEN T 'S WEEKLY PROGRESS REPORT

Full Nome......................................... ..lab. Sec. # ...... Date.....
Plen.se refer to your study record on your TEEM PROGRESS SHEET AND STUDY LOG
in answering the qiiestions below; place a check mark on yo\ir progress
sheet to indicate the reporting of the information which you plAce on this
report.
Question #1: How many pages of the assignments in the text have you
read since last reporting?.........................
-pages.
Question

i»?:

How many pages of the assignments in the lecture syllabus
have you read slnce last reporting?...............
pages.

Question # 3 : How many pages of the assigned library readings have
you reed since last reporting?..................... ..... pages.
Question

How many of the assigned laboratory guide "Study
Questions and Problems" have you completed since
last reporting?
...................................

pages.

Question #3: How many hours have you studied for this course
since last reportlng (excluding class time)?......

hours.

Note:
In order for the research involved to be of any value it is nec­
essary for you to cooperate in making your data as accurate as possible.
Thanks.

113

APPENDIX III
CLERICAL AND TABULATING MATERIALS

FLOW CHART OF THE CLERICAL AND TABULATING PROCEDURES

1.
Questionnaires were deposited,
with the investigator "by the
cooperating instructors.
?.
Questionnaires were sorted and
filed alphabetically by the
investigator

7.
Clerk: calculated the percentages
of assignments completed and
average number of hours oer week
reported studied; each calcula­
tion was checked.

.

8
Calculated data, with data from
other sources, was transferred by
the clerk to a punching schedule.

3.
Stens 1 and o resulted in a
composite file of all question­
naires in the survey with all
onestiormnires for each res”ordent filed together.
U.
Dfitf ,/t-ri. transferred by a cleric
v.uh.t in entries as the investigator
cnecked against the questionnaires.
5.
Data sheet was verified by clerk
rer-.<•*,in -T entries as t: e investigator checked nginst the nues—
tionnnires•
b.
Steps U and 5 resulted in a ver­
ified composite data sheet for
calculations of percentages.

9.
Data on the punching schedule was
verified by the investigator read­
ing the entries against sources
of data on the composite data
eet .
10 .
Steps 7 through 10 resulted in a
'unching schedule verified for all
data, to be analyzed.
11

.

Key punch, operator punched the
Hollerith cards from the punch—
ing schedule.
1?.
Pxinched cards were verified by
the investigator reeding from a
data sheet printed from the cards
and the clerk checking the rmnching schedule.

APPENDIX III. PLOW CHART, Con I'd.

13.
Stenos 11 and 1? resulted in
verified rmnch cards.
lU.
IBM comr'UtiniT machine operator
obtained data for substitution in
correlation formula; machine checked
at each step.
15.
Investirator computed the corre­
lations from the IBM data; each
computation double— checked.
1^.
Investip-fltor computed and applied
Student t-test for sienificaace of r.

115

APPENDIX III, Cont'd.
SAMPLE HOLLERITH CARD

—«» — « • —
iO ** «■ •9
& u CO **
■
*
•• *
o •- CO 4 «u 4
J
»
CO «D 09 «
3 co •* c» «•
g C
O
CO ~-

9* «_!» 49

cn

•• cn

CO ••
<o

CO ••

■
—i

CO 5
CO — CO ft

ft-*
-O

k

o» k

-O

CO a

CO k

---->

oo

■
—
J

i; CO ft
T. CO ~z

C
O

<n

ft

o» O
.
C
O

— CD ft
— cr* ft

ft CO ft

C
O •

k CD ft
& fO

CO
ffO

O
ff C
rc CO

C
O C
O H

*
—
1

03

CO :t

—
J

cn

C
O y
O a
v» C

—
*

03

a
y

cn
03

co

O
O k

;* CO ft
IK 1. 3
to

k
-o’

4 -

a
u

91

<73 a
—
J J* tn k
—
-j 2 03 k

cn
cn
cn

k

k

03
JJ on

a

a

—i 2

03

ts

«n

09

k

--4

2

03

k

cn

oo

»

--4 S

03

b.

cn

oo

k
“

—
J

03

k
ft

cn

09

K

t- CO c

oo

C

U1 i

oo

t

03

■
5 —t

09

x.

’-—
: ft
«-J ft

03

ft

e*-.

c

—
-4

ft

k

J ■

cn

J.

ft

* --.

4'

cn

t.
*

cn

—
~i

ft

03

ft

O'

*
ft

cn

— i

03

£

' '3

IT
v Oo ft
O 2
i: C

<93 «
09

------ sc
ft

e

cn
tn

c >

c-

L3 5
.

<73

til

ft

sc

—
J i

<r

o

J

C
D a;

~-~4 ft

(/.

ft

O
O k
ft
oo ■

4 £

rr» ir
O’ h.
03 ft

\c. vO ft

09 ft
ft. CO X
C
O 5? 09
'.’J -at C
O ?

-—
1 '<Z
_
_
__ s
■

r

'O

S

------ ft

c

X.

*

O
-■ C
■
C
i ft, ro ft

ir. J3
*

S

r; Cl
C
O J1 oo

O ft* 09 *
i? C
ft C
D ft
"i . r
oo

*.
z: -in

oo
-.

•O'

.;

<
*•
C
C

-j i_.J
to
.«.

J ',:

-i

n

cn

<j

c •
03

09

<73 s:

«yi

*-

..»

4k

r

0»

cn
cn

ft

ft
• •> jj

«.♦•
<n

*
;

mr.
mm

cn

<
*
-■

cn

A

mm
mm
mm
mm
cn

C
7

O'

r..

*

' -i

W
4
,1

4>

c <
<-

<o-

cn

A.

y

CD ft

74

• A

y

cn

k

-9-

c_

m

cn

a

7k

ft

c > «4
C
D 4
ft
4

ft
£
£
ft
8
ft

<T
a

ft
ft

o

JS

;]

f

—
T
> ft

c

C
~
> *
ft

r
ft

•ft

<
-» *
e .» ft

3
ft

_3 ft
>
.
«T - - oi C
«
*
■ r .3
ft
—■ 40 o
-.3 f»

ft
3

i

•:

•j

~-----

►
o £

-*

ro

t-

y.

- -

r. i
'3

<
3

3

►
.j r
r_

x
r -> ft
rsj y
;
- .■ ft
.:

----- J
T
! c~
.. . ft tr.
o

-•
—
i

■m

ft
ft

8

ft

ft

ft

s
2

r
s

5
ft

- *

c=» 5

ft

—
* £
“

CJ

s

..
rO a
ro

J3

• D 3;
ft

-ft

9

?S

—
D s
<
r~i a
tr

. i

■
*4 v-i
4-

91

k

ft

•O

O.■

tn

c. * 4
CD r
C
_3 4

ft

3
4

t~»

<iJ c

4-

4
4
,

*-

ft
ft

if

c
T
C
.3

Jv ft
-

•--

c

4*

44 £,
^
3 -D.
K. s
8

m.

m

-J

ft 0 '
4
. —
C
T </ 09 ft'
i
O
O a
f_J
^.4
O
C
O f, O
—
4 a

a

t

03

or

c*.

c.

•----- z

«a> a
<n
4*0 •3

<~7 ft

o

«.r

C
O

o
■
“
,

-^4 ft

' -J ."

ft'
C
D -

<

C...

■
----- ',-

3
ft

ft.
ft

r 3 •*
r. o

C7.

■
----- 3.
- 4 •

r

U
V 4«
4*
i,

ft

if

c > £
(_
_J j.
4k

Si

ft

. j

- 1

r

3
8
if

ft

<7»

*

ft

o

ft:

£

3
ft

£

-o

-,4

C
.

ft

- -

ts

C.3 ft

J

xz

c«
■
,
Cn

—

,

<
. .

cn

03 tt
<73 a

*.

.*
♦

c_n

m

♦
*

r

e.n

X

co

m

t-

- -

r

C3 ft
CJ ft

ft
— si o
■
. <
— *
—
> J5
- .
c*
--

ft
V

—
ft . 3 ft
~* S
----- * C» ft

*
•

».

3
t*
ft

J o
:-

ft

c
4-i

r:

c

4
.' Or.

y

s
ft

*

<43 V

m

r S
ft

£

,
C“
1 •*

CT j

r

-~4

I-

O
'? a

%

C
f> ft

—
»

B
K> r

t— 4
oo
C-

*

if

V
«

C
O ft

C
D

- *

r_

c-> fe

s

j.

ro

-o

♦ •o

C
O

ro

ft,*
<
•
ft t .» a
<o

cn

--

4— ft O
O

-C
'.

w

.--

- • 0
- ♦ft- O
ft o
— S

*

ft

03

2

4

ro

<71

V

<73

T
3
—
* ts C
5
t
----- 5T 70 *
O
—
_3 ft
£ V

ro

-

O £
ft C
4 cr- 4

—
4 ft;
rao •S
-4 ft
» —
J ■
09 V
----- 1
C
■
------ y*
CO

y

c-.’

cn

c>

oo

ft

cn

toO ;4

*i
r.„> ft

ft

7».

<7* k

4

o

ft

k
■
ft

cn

03

V
<7> e
03 "ft

ll

C
O

—
4
£

-Z—

k

k
r . B
►
_, ft

ft

------

ty» 6

~

i'
ft

<
1.1

- .

4^

k

r

t
O £
5: C
A cr. f:.

c«

k

«-f~
> .*.
C
O -

t

03

4—

cn

oo

t-

y
ft

ft

L, O

y

ft
a

k

—

L.

s
a

ft; cr*

—

t;

a
3

r

tj

> ft

19
«*

a
m

c>

—

k

4

O ft
—
■ ft; C
—
. a <=» ft
.
CD

4

---•

^

tA' U
o
. «

<:> ft

•9

m
m

4

-----

.•

*ft c •
<7 7(
■
ft O
O

rt

O

m
m

--3 >*
►
-> ■
KJ li

ft

f O rt
r j

4^

o

c-> ft
T
-; T3 ‘ J
ro *
_
*

1
L.
■
ft . o ■
-

cn

oo

oo

'-..3

♦•-

ft

i o

C
*.

j

-

r •

9

m
m

ft"
* S c . ft
ft c=>
—
» ft < 7

— —
k £

o;

—
*9

-• CO *-

m
—
- m
—

»J

t > X
j. ro .;
m ►
o

—
«*

m
—
• — c=> m

•9
x

co

•-

C
D y
cr.

ja.

c~>

.ft '7
...
/ft

C

<n

O ft
k C
O S
fc C
C
O ft
k

oo
09

7*.

:t

'~~r

KJ 0
►
sJ

oo

cn

an

Kj m

t-o

ft tli

,i

O
K C
e U3 “

C
O ft
k CO ft
ft
40
k

ft
4^
7-

foJ <•

C
-J

ii

—«

C
O a

if’ » '

ft

03

co ft

oo

cn

_»’
ft <
V
*4 CD ►

C
O a

ft

-----

a

^
4» o
—
» •4 O
_
_
__ <
DO

K1 ------------

m

— -D
*i-

ft

03

r* co
O ft
s
mC
C
O ft
*

cn

1-o m

—(k
»
•»

«•

T
» i- «
_
/■ r
3J C
cn r- cn
*
—
O
03
cn

— _
__ —c »
r* o
^

«47 4,
*33 4.

-* CD
— til

-3

cn
iL OT
—03 ft cn
CD r~l cr> ft

CD — •sa

*- C9 4.
*
4 *k
.> »
»• 3 «.

«> 4^.

— cn

cn

•

X

•3 91

cn

o» ••

♦ CO ft
CO a

«» o»

•9.

cn

-

•» CD

M

•- Cn
••

-o

*•

tt CO
O y
V C

*

j

*4

—*

co

co

;

<yt
•—
j •4. <n

«»■

— «-n

» *- cn
•*
>
-« »* <X
Cn
o*

•

t
:

r

—
* —
• 91

.->
t 3

s
3
rf

.-j
w
y

a
4

4

o

’ 3
•O

ft

<
.=
> 3

k

ft

o

a

r

ri <
■.< ft
^

►
_
i
►
o a

- * ...
ft

' -

k

«- • k
cr-* 3

•— K •-9

3

k

k
S

I

116

APPENDIX III, Cont*d.
DATA CARD USED FOR THE COMPUTATION OF THE CORRELATION
COEFFICIENTS IN THIS STUDY

<

_____ Front Side______ __________ __________
1
x
5
X
1
y
5
Data for computation of Pearson product-moment of r^y.; (Hollerith Card)
Instructions to IBM computing machine operator: Pull all cards having
an X punched in last column of field being used for computation.
x data in columns_____

Decimal point______ (

Y data in columns______

Decimal point

N51 Number of cards having x and y data.
^

__________________________ V

£x^s

______________________

_______________________ t =___________________

_____________________ Reverse Side________________
_____________N_( X Y ______
/

cn

_- T i x)^]

L _ _________________) -

(*X) ( £Y)________
[FIy5

-

t T y ) 2T~

(_____________________ >

J --------------------------------------------------------

/

to____

(_____ to____

117

BIBLIOGRAPHY

Abelson, Harold H. The Art of Educational Research.
Book Company, 193333^ pt>.

New York:

World

Allison G . , and A. Barnett.
"Freshman Psychological Exam Scores as
Related to Sizes of High School,” J o u m a l of Applied Psychology,
?U:65I-65?, October, I9U0 .
Almack, John C. , and A. R. Lang. The Beginning: Teacher.
Houghton Mifflin Company, lhPS.
U 7? nnu

Chicago:

Anderson, I. H . , and W. P. Dearborn.
"Reading: Ability as Related to
College Achievement," Journal of Psychology, 11:337-396. April,
19U 1.
Armsby, Henry H. "Colleges Teach and Practice Democracy," Journal of
Higher Education, U:196— 199, May, I9 U?.
Ashmore, B.
"High School Teachers’ Marks as Incicatcrs of College
Success," American Assoc la tlon of Col lege Registrars1 J o u m a l , ?1:
? 19-230, January, 19^4-6.
Avent, Joseph., E. Beginning Teaching.
nessee, 1927*
599 pp.
ville:

Knoxville:

University of Ten­

Excellences and Errors in Teaching Methods.
Joseph E. Avent, 193^*
5^9 xrp.

Bagley, William C. , and John A. Keith.
New York: The Macmillan Company,

Knox­

An Introduction to Teaching.
Uqo oo.

1 "<PJ»

Bagley, William C., and Marion E. MacDonald.
Standard Practices in
Teach ing. Nev- York:
The Macmillan Company, 193&.
1^9 op.
Bain, Reed. "Stability in Questionnaire Response," American Journal
of Sociology, 37:^U5-U53, November, 1Q31.
Barnard, J. D. "Groun Process in Science Education," Science Educa­
tion. 33:^67-?69. October, 19U9.
Bird, C. , and D . M. Bird.
Effective Study Hnbi ts.
ton Centura'- Co., Inc. , IQU5. ?75 P7> •

N*u York:

D. Apple­

Blacdiurst, J. Herbert.
Pi rectec O b s e r vat ion and Sunervisod Teaching.
Boston:
Ginn and Company, l^^^.
T ' O pp.

118

Board of Exa.~ir.ers, Michigan Stnte College.
Comprehensive Examinations
in a Program of General Education. East Laris inf*: Michigan State
College Press, I9U 3 . I65 PP»
Bossing, Kelson L. Progressive Methods of Teaching in Secondary Schools
Boston:
Houghton Mifflin Convoany, "lhu?. 779 pp.
Breed, F. S.
Mhovr Mehe Teac'-.ins Democrat ic?M SchoolRevlex?,57 •531"*539»
December, lpLp.
Brino, William G. Pirectins Study A c *irltios.
Doran and Company, Inc., Ipp7.
I } 7 T'tr.

New York:

Doubleday,

Broom,
S.
"The Importance of Reeding for Col', ege Study,11 Educe, tional
Adrr.ini strotion and Supervision, 20:1?Q— lh?, March, lppu.
Brubacher, A.
Corrmany,

R.

Teaching: Profession and Practice. New Yor^:

1~"".

Century

*01 no.

Bucki r.ghar.,
B. R.
"The
^jestionnaire,H
Journal
of ^Educational
Research,
_ ' . r'
_ 1 -c__
_
Jt
“■ ■ “ |p“
^
1■— 1
.'ur.e , i^ o .
Burton, W. H.
P.

S’uuervi s ion ar.d the Imo ro v e.nen t of Teaching.

A y r l e t on a n d

_______ .

The

Company ,

Ip"'''.

New York:

3 11 y .

Guidance o f Learning Ac t ivi 1 1 <=>1

New York:

P. Annie—

ton-Century Co.T.r.any, 1p'*
Butler, Frank A.
The Iiu~ rove
,t of Teaching in the Second--.:
hioa.-o:
The Yr.iversity of C-.ioago prog'.,
n:
S''cc e ss

hi h. School

SC.'OO^S.
ions

DC.rr 0
iarr

Willi ar.
ve r sity:

a:
S tar.f or: Univers:
.iii

c .

Lesson

Stonfcrd Ur. L-

C-nvan, Ruth 3.
MTr. e
ue ,tionnaire in a Soci'lo^ic 1
American J u m a l o f i'.'C.ologo , J * • j '1— 7 • La
.

o ear:

*oj ec

■les, P. C.
"College Performance of To. n— ^uarter hiho School Gradu­
ates,** J o ur. a 1 o f Ed-cat lonal P B y c h o l o g - , 3p:7'?— pi. Feb rear*.*, iphW. W.
Conoanv,

C ’irric'u m

Ipou.

750 00

Construct: ~n.

feo Yor.::

The Macmillan

119

C l a r k , E. H.
"High S c h o o l S t a n d i n g a n d C o l l e g e G r a d e s , " A m e r i c a n A s s o ­
c i a t i o n o f C o l l e g e R e g i s t r a r s * J o u r n a l , 1 6 : P 7& -279» A p r i l , 13^1
C la r k , E. L.
" H i g h S c h o o l A v e r a g e V e r s u s H i g h S c h o o l C l a s s Rank a s a
M e a n s o f P r e d i c t i n g C o l l e g e G r a d e s , ” S c h o o l and S o c i e t y , 3 ^ : 7 - 5 ~
7 6 6 , D e c e m b e r 5 * 1'93^«
C o h e n , M o r r i s R . , and E r n e s t Nagel.
An I n t r o d u c t i o n t o L o g i c and S c i e n ­
t i f i c M ethod.
Ne.~ Y o r k :
H a r c o u r t , B r a c e and Company, 1 9 3 <:i» ^ 7 PP»
C ole,

L. W . , and J. M. F e r g a e o n .
S t u d e n t s * Guide
N e w York:
Rin eha rt and Company, Inc., lpll*

to E f f i c l e n t S t u d y .
7'G PP.

C ole,

L u e l l a . _ T h e B a c k g r o u n d f o r Col l e g e T e a c h i n g .
a n d R i n e h a r t , I n c . , 19-tO.
t>l6 p p .

New YorK-:

C olvin , Stephen.
I n t r o d u c t i o n to High S c h o o 1 T e a c h i n g .
M a c m i l l a n Company, 1 9 1 7 .
£51 pp.
C o n d i t , P . M.
ifica tio n
3 3 5 . May,

Farrar

New York:

The

"Trie P r e d i c t i o n o f S c h o l a s t i c S u c c e s s by Means o f C l a s s ­
E x a m in a tio n s," J o u r n a l o f E d u c a tio n a l R e se a rc h , I Q ;371—
19?9.

C o r e y , S . M.
" S ign ed V ersu s U n sig n ed Q uestionnaire-:-,"
c a t i o n a l P s y c h o l o g y , ^ : l U h —l b 3 f May, 1 9 1 7 .
Crawford, A. B.
N ew Haven:

J o u r n a l o f Edu—

Incentives to S t u d y : A Sur v e y o f SJtudenjt O p i n i o n .
Yale U n i v e r s i t y Press, 1°?9.
19^" po. *
~

C r a w f o r d , C. C.
Tr e T e c h n i q u e o f R e s e a r c h i n E d u c a t i o n .
U n i v e r s i t y o f S o u t h e r n C a l i f o r n i a . lPPft.
jpO o p .
_____________ .

Depository,

How To T e a c h .

193^*

5H

Los A n g e l e s :

L o s Angeles:

South.ern California Book

PP«

C r o x t o n , F r e d e r i c k E . , a n d D u d l e y J. C ow de n.
A p p l i e d Gener a l S t a t i s ­
tic s.
New York:
P r e n t i c e - H a l l , I n c . , 1939*
9^*4 pu«"
C u r e t o n , E. E.
" N o t e on t h e V a l i d i t y o f t h e ACE F s y c r o l o g i c a l
J o u r n a l of A p p l i e d P s y c h o l o g y , ? 3 I 3 ^ h —3 ^ 7 . A p r i l , 1939*

Exam,"

D n v i s , R o b e r t , and E dw in L. B a r r o w .
" C r i t i c a l S t u d ; / o f trie . Q u e s t i o n ­
n a i r e i n E d u c a t i o n , " E d u c a t i o n a l Admi n i g t ra t j o n and S'.iperv i s i o n .
'"*1:137— l^h, February, 1935*
D e a r b o r n , Ned H a r l a n d .
An I n t r o d u c t i o n t
i l e t o n a n d Company, 19 ^ 5*
337 PP*

T eaching.

New Y o r k :

D. An­

120

Douglas, H. R. M o d e m Methoda in High School Teaching.
H oughton M i f f l i n C o ., 1 9 2 6 .
5 ^ PP*

Boston:

Dressel, Pnul L. ’'The Effects of the High School on College Grades,”
J o u m a l o f E d u c a t l o n a l P s y c h o l o g y , JO : 6 1 2 —6 1 7 » N o v e m b e r , 1 9 3 9 Dwyc-r, P. S.
"Some Suggest I or.s Concerning the Relationship between
Size of High School Attended and Success in College,” J o u m al o f
Educntlonal Research, 32:271—281, December, 1933.
Ellis, Max S., and Arthur J. Riopieee, "An Efficient Punched-Card Meth­
od of Computing X, X^, XY and Higher Moments,” P3ychonietrika, 1 3 t
79-SB, June, 19U*.
Ewer, B. C.
College Study and College Llfe.
1917. 228 pp.

Boston:

Ezekial, Mordecai.
Methods of Correlation Analysis.
Wiley and Sons, Inc., 19^7*
531 pp.

Richard G. Bodger,

New York:

John

Finch, F. H . , and C. L. Nemzek.
"Prediction of College Achievement
from Data Collected during Secondary School Period," Journal of
Applied Psychology. 18: U^it— it-60, June, 1939*
Fisher, Robert P.
"Signed Versus Unsigned Personal Questionnaires,”
J o u m a l o f A p p l i e d P s y c h o l o g y . 3 ^ s 2 2 0 —2 2 5 # M a r c h , 13^-6.
Fontaine, E. Clarke.
Ways to Better Teaching in the Secondary Schools.
Boston:
Ginn and Company, 1^28.
°71 p p .
Ford, Frederick A. The Instructional Program.
Hall, Inc., 193??. U 58 pp.

New York:

Prentice—

Foster, Herbert H. Principles of Teaching in Secondary Education .
New York:
Charles Scribner's Sons, 1P21.
3^7 pp.
Furfey, Paul H., and Joseph F. Daly.
"Product—Moment Correlation as
a Research Techninue in Education," J ournal of Educa tlonal Psy­
chology , 29:206-211, March, 1935*
~
Gates, A. I., and Jessie La Salle.
"The Relative Predictive Values of
Certain Intelligence and Educational Tests Together with a Study
of the Effect of Educational Achievement Upon Intelligence Scores,”
J o u m a l of Educational Psychology, lit: H I ? —5 3 9 , June, 1927.
Gerberich, J. B.
"Study of the Consistency of Informant Responses to
Questions in a Questionnaire,” Journal of Educational PsreholOjgy.
3 8 : 2 9 9 - 3 0 6 , M a y , I 9 U7 .
‘

121

Gerberich, J. B . , and John M. Mason.
"Signed Versus Unsigned Question­
naire fM J o u m a l of Educational Research, U2:122-l2b, October, I9U 3 .
Gillis, P. M.
"Correlates of Intelligence in College Students," School
and Society, 3l4G
270, August 22, I231.
Gi Luo r e , Marquis E. Exemplifying Go ~d Classroom Methods and Procedures♦
Boston:
The Christopher Publishing House, 1931*
282 pp.
Goetting, 1!. L.. Teaching in the Secondary School.
Kali, Inc., 19^2.
519 op.
Go^d, Carter V. Plow To Do Research in Educe 11on.
and York, Inc., 1329.
293 pp.

New York:
Baltimore:

Prentice—
Warwick

Gordon, H. P.
"Study Habit Inventory Scores and Scho larshin," J o u m a l
of Applied Psychology,11 23:101— 107, February, 12-1.
Greene, E, G.
"Effectiveness of Various Rates of Silent Reading of Col­
lege Students," Journal of Applied Psychology, 15:?lU— 227. April,
1931.
Roll — Quest, A. L. Supervised Study .
1916.
^73 p p .

New York:

T'ne Macmillan Conoanv,

Hamilton, E. P.
"T’
ne Textbook from the Publisher’s Point of View,"
The Wiley Bulletin, XXXV, #1, Spring, 1952.
Hartman, G. W.
"T'ne Interview as a Research and Teaching Device,"
J o urr.a 1 o f Aon lied Psychology, 17s2lo— 211, April, 19^2.
Hawks, L. J.
Certain Relationships Between Scholarship in High School
and lr. College. Baltimore: John Hopkins University, 1231.
B3 pro.
Keer, Amos L.
Steps to Better Teaching.
Conroany, Inc., 1937*
313 op.

New York:

Holley, Charles E. High School Teachers1 Methods.
rard Press, 193751U pp.

W. W. Norton and
Champaign:

The Gar­

Holzineer, Earl J.
Statistical Methods for Students ir. Educ-tion.
ton:
Ginn and Company, 1923. "372 pp.

Bos­

Hubbard, F. W.
"Questionnaires," Review o_f Educe ti onal Research. 9 :50?522, December, 1?3?«
:11,

Clark L.
"Tr.f- Co rr ela tio n Coefficient and its P r o g n o s t i c V a l u e ,"
J onrr.al of Ed :ca t i^nal R e s a o rch , 1 r : 7^7-3 33 , May, I- 2 7 .

12?

Humber, W, J, "The Relationship Between Rending Efficiency and Aca­
demic Success in Selected University Curricula," Journal of Educatlonal Psychology. 3 5:17~2b, January, 19^9.
Kelley, T. L,
Statist!cnl Method.
1 -?'-. 700 pO.

New Yor>:

The Macmillan Company,

_____________. Scientific Method: It3 Function in Research and
tion. Colunbus:
The Ohio State T’niversity Press, H P q .

Educa195 "on.

Kilpatrick, William H.Foundo11ons of Method.
Conroany, 1925*
3^ '
D~>»

Macmillan

N^w York: Tne

Knudsen, Charles W
Evaluat ion and Irrinrovenent of Teaching.
Doubleday, Doran and Comuany, Inc. , 193?. 53^ on.
Koos, L. 7.
The Questionnaire in Education.
Company, 1928.
173 PP.

York:

New York:

Tne Macmillan

Krathwohl, W. C . "Effects of Indastrious a_nd Indolent Work Habits on
Crade Prediction in College Mathematics," J 0um a l of Sducational
Research, ^.3:32— Uo, September, lqbh.
Krech, David, and Richard S. Crutchfield.
Th_eo_ry and Probl ems of Social
Psychology. New York:
Mclraw-Hill Book Co., Inc., 19^-3. ~ o n .
Lancelot, W. H. Permanent learning.
Inc., 1929.
lQf op.

New York:

John Wiley and Sons,

Landry, T. M. "Relative Predictive Value of Certain College Entrance
Criteria," Journal of Experimental Education,
PrO, March.
1 oir
"
MacPhail, A. H. "Q, and L Scores on the ACE T s-chological Exam,"School
and Society, 5oS?L3—25^-1 Septmeber 19» 19^?•
Maxwell, C. R. , and W. C. Reusser.
Obg erva tion and Direct ad Tencr inf;.
York:
Prentice-Ho 11, Inc. , 1°39.
Gy U
Mo-*, M. A.
"Predicting Academic Success," J o u m a l of Educational Psychplogy, lb;Loq-LLO, September, 1923.
k_
Mclehee, W.
"Freshman Drnies and th.e American Council Fsyc’
nolomi cal
Exam," Scho ol and 5~ ci ety, ^7:222— p?L, February 12, 1932.
Mendenhall, R. M., and R. Warren.
" Comp *.t inm Statistical Coefficients
from P ^inched Cards," J 7 irnal of Educat ional Psy chol pay, °1:~~— £2 ,
January, 11J.0.

(

123

M i l l e r , H arry L,
D irectin g
19??377 p p .

Study,

New Y o r k :

C h arles S c r ib n e r 's

Monroe, W a l t e r S ,
D i r e c t i n g L e a r n in g i n t h e H igh S c h o o l .
Do■iVleda.y, P a ^ e a n d Company, 1 9 ? 9 *
577 P^>.

Sons,

New Y o r k :

___
•
D» S . S t u l t " C o r r e l a t i o n A n a l y s i s a s a Mea ns o f
S t u d y i n g C o n t r i b u t i o n s o f C a u s e s , * 1 J o u m a l o f E x p e r i m e n t a l Edu­
c e t i o n t 3 : 1 5 5 - 1 6 5 , May, 1 9 3 5 .
" N o t e on t h e I n t e r o r e t a t i o n o f C o e f f i c i e n t s o f C o r r e ­
l a t i o n , 11 J o u m a l o f S d u c a t i o n a l P s y c h o l o g y , p 71 551**557'» O c t o b e r ,
193^________________ , e d .
E n cyclop ed ia o f E d u ca tio n a l R esea rch .
The M a c m i l l a n Company, 1 9 5 0 .
15?0 pp.

New Y o r k :

M o r r i s o n , H e n r y C.
""he P r a c t i c e o f T e a c h i n g i n t h e S e c o n d a r y S c h o o l .
C hicago:
U n i v e r s i t y o f C hicago P r e s s , 1 9 2 6 .
6^1 p p .
M u r s e l l , J-i:nes L.
Suc cessful Teach I n g .
Company, I n c . , 1 9 ^ 6 .
33S p p .

New Y o r k :

M c G r a w- Hi l l Book

N o r t o n , J . K.
T he Q u e s t i o n n a i r e .
N a tio n a l E d u cation A s s o c ia t io n
sea r c h B u l l e t i n , V I I I , ffl, Jan u ary, 1930.
N utt,

H u b e r t W.
Tr: e S u p e r v i s i o n o f I n s t r u e 1 1 o n .
M i f f l i n Company, 1 ^ 9 0 .
977 pp.

B oston:

Re­

Houghton

O l s o n , W. C.
11 Dae W a i v e r o f S i g n a t u r e s i n P e r s o n a l Data R e p o r t s , "
J o u r n a l o f A p p I i e d P s y c h o l o g y , 2 0 : UUp—U5 O, J u n e , 1 9 1 6 .
Crr,

H. K.
" C o l l e g e A c h i e v e m e n t ; c o m p a r i s o n o f t h e r e c o r d s made i n
c o l l e g e by s t u d e n t s f r o m f u l l y a c c r e d i t e d h i ~v s c h o o l s -r if. 1 <• : t
c f c i n d e n t s h a v i n g e q u i v a l e n t a b i l i t y , from 2nd , and 3 r d c l a s s
h i g h s c h o o l s , 11 J o u r n a l o f Ed u c a t i o n a l R e s e a r c h . ^ 7 : 3 5 3 “ 3£>^» J«-n—
• i a r y , l r-'^9.

P a r k e r , S a m u e l C.
and Company,

Metho d s o f T e a c h in g
19?0.
5?9 po.

in High S c h o o l s .

Boston:

Ginn

Per -tn .a n, J o h n G. , e nd I s r a e l G r e e n s p a n . MR e l i s . h i 1 i t y
of a
Q u estion n aire
on t h e S u p e r s t i t i o u s B e l i e f s o f E l e m e n t a r y S c h o o l C h i l d r e n , 11 J o u r n a l o f A b n o n r . a l and S o c i a l P s y c h o l o g y , 3 r; 59 0 S —9 9 1 , J u l y , I 9 9 5 .
P o w e r s , S . R.
"An E v a l i u i t i o n o f S c i e n c e T e a c h i n g i n S e n i o r H ig h S c h o o l , "
Th e S c i e n t i f i c M o n t h l y . V o l . 7 ^ , # 5 , 2 7 3 - 2 7 5 , May, 1 9 5 %

l?U

President's Commission on Higher Education.
Higher Education for American
Democracy. New York:
Harper and BrothersT 1§V7.
Vol I* 103 PP«
Raup, R. Bruce, George E. Axtelle, Kenneth D. Benr.e, and Othan Smith.
The Inip.rovement of Prac- tical Inte1ligence. New York:
Harper and
Brothers, 19^7•
3^3 t>u.
Reagan, G. W.
Fundamentals of Teaching.
Company, 193?.
35^- Pr'*

New York:

Scott , Fores mar and

Reeder, C. W.
"Study Habits," School and Society, Up:^il3~^15» September
°1, 1937.
Reeder,

Edwin H.

19?9.

S i m p l i f y i n g Teach.ing.

N e v York:

L ai dl aw Brothers,

19? pn.

Reeves, Charles E.
Standards for High School Teaching.
D. Appleton-Century Co., 193>°. 55^ '
D'°»

New York:

Recun, D.
"D o r s High School Rank Determine College Success?"
An e rl can
Association of College Registrars * J oum a l , l6:?99“3Ci5» April, 19^1.
Rigg, Melvin.
"The Relation of C o l l e g e Achievement Tests to Grades and
to Intelligence," Journal ojf_ Educational Psychology, kO:397“ ^00,
Hay, 1939/
Risk, Thomas M. Principles and Practices of Teaching in Sec n-.dnry
S c:;oo 1 s • New York:
American Book Co., 19^!l.
I P S pn.
Ruediger, William C. Teaching Proceduros.
Company, 193".
^77 pp.
Russell, Charles.
T *=-c' 1r.g frr T o m o rrow.
1937*
^77 pr.

Boston:
No- Yrrk:

Houghton Mifflin
Prr r.tice-Ha 11, Inc.,

P-yens, D. G.
"Some Observations Concerning the Relationship of Time
Spent at Study to Scholarship and Other Factors," Journal o£ Edu­
ce 11onal Psychology, 30: 37?~377 * May, 193‘7.
Saupe, Mildred W.
"Size of High School as a Factor in College Success
of Average and Superior Graduates," Amer ican Associa.tlon of College
Regirtrars1 Journal , "LJ
, October, 19&1.
Schneider, G. G . , and R. F. Berdie.
"Educational Hierarchies and Scho—
lantic Survival," J o u m a l of Educut ional Psychology, 3311CI7— r*08 ,
March, 19^?.

125

Seder, M.
"Reliability arid Validity o f the ACE Psychological Exam,
1 9 3 ^ e d . , H J o u r n a l o f E d u c a t i o n a l R e s e a r c h , 3 ^ : 9 0 —I d * O c t o b e r ,
19UO.
S e g e l , D. , a n d J . R. G e r b e r i c h .
" D i f f e r e n t i a l C o lle g e A chievem ent P re­
d i c t e d "by t h e A m e r i c a n C o u n c i l P s y c h o l o g i c a l E m m , " J o u m a l o f Ap­
p l i e d P s y c h o l o g y , 17 S & 3 7 —r^ 5 » D e c e m b e r , 1 9 3 3 *
S e l l s , S. B.
" O b s e r v s t in n a l M ethods o f R e se a r c h ; ^ues t i o n r . a i r e s , M
R e v i e w o f E d n c a t i o n a l R e s e a r c h , 1 8 1 ^ 3 ^’” ^3 8* D e c e m b e r , 1 9 ^ 9 .

Seyler, E. C. " Value of Rank: in High School Graxiuation Class for the
Prediction of Freshman Scholarship,” American Association of Col­
lege Registrars1 J o u m a l . 15:5-??* October, 1939*
________________ ,
" T h e Value o f Rank i n H i g h S c h o o l G r a d u a t i o n C l a s s f o r
P r e d i c t i n g S c h o l a s t i c A c h i e v e m e r t i n C o l l e g e , ” Ame r i c -‘n A s s o c i a t i o n
o f Co1 1 p e e R e g i s t r a r 3 1 J o u r n a l , l 6 : l ? 3 —13 3* J a n u a r y , 19^ 1 *
S h n t t l e w c r t h , F . K.
"Study o f the Q u e s tio n n a ir e T e c h n iq u e ,” J o u rn a l
E d u c a t i o n a l P s y c h o l o g y T P P : 6 5 ? —9 5 S , D e c e m b e r , 1 9 3 1 *

of

Sm ith, F r a n c is F .
" D ir e c t V a l i d a t i o n o f Q u e s t i o n n a i r e D a te ." Educa t i o n a l A d m i n i s t r - : t i ^ n a n d S u o e r v i °< io n , P I : 9 6 1—9 7 9 * N o v e m b e r , 19 7 ? .
S t 1 t h , T. V. , and Ed ua rd C. L i n d e n a n .
The Democ r a t i c Way c f L i f e _ .
New
York:
The New A m e r i c a n L i b r a r y o f W o r l d L i t e r a t u r e , I n c . , 1 9 9 1 .
199 on.
S t e w a r t , W. F .
M ethods o f
verr’to Press, lc90.
3 torn.:: a n d ,
M ifflin

rtir. J .
Company,

Good T e a c h i n g .
Pp~i pr>.

Columbus:

P r o g r e s s I v e M e t h o d s o f T e a ch i n r .
I 0- 7 .
7 C1 PP*

S y m o n d e , p e r c i v a l M.
"M ethods o f I n v e s t i g a t i o n
ar.d S o c i e t y , PL*
j u i y 71, 19?6,
T h a y e r , V. T.
Conmany,

Th e P a s s i n r o f
l^PS.
7 3 1 Vr'»

the R e c it a t i o n .

T).o.y?is , F . W.
P r i n c i o l e s and Tec' n i c u e s
M i f f l i n Company, 1 9 ? 7 .
^10 pu.
T h o r a b e r f r , L. H.
and S o c i e t y .
Touns,
T.r

The O h i o S t a t e TTn i -

of

o f St ud y H a b i t s , "
Boston:

T e a c h i ng ♦

' ' C o l l e g e S c h o l a r s } . i p and S i - e
; j : 1 8 9 - 1 9 ? . A v * u s t 9 , 1P?U.

B oston:

o f High

D.

Houghton

Scho o l

C. H e a t h a n d

C hicago:

S ch ool,"

Houghton

School

Herbert A.
" V a l i d a t i n g the Q: s t :onr.ai re M e t h o d , " Journ al o f
He search , Pt 1 9' — IP.Q, A .■" ;s t— Sen t embe r , 19?" .

l?6

Trebilcock, W. E.
"I*Ian:r of the Lowest Third of our Graduates are Col­
lege Material*" £leering ^ ° • 1 ? ! ,
May, 193^»
Triggs. Frances 0.
"Reading at the College Level," J o u m al of Higher
Education , P C ’i f ^ - l C , F e b r m r y , I9U9 .
Tschuprow, A. A. Principl es of the L'ath.emnt leal T}.eor\~ of Correl ati on.
New York:
Nordeman Publishing Co., Inc., 1939*
19^_ pp.
Tuttle, G. P.
"The Predictive Value of Rank in High School Graduating
Clasp," Bulletln of the American Association of College Reglstrars,
18:111-118,‘January, 1937
Wade, N. A.
"Comparison of the Time Spent by First—Year Students and
Eroectfd by Teachers in a State Normal School," J o u m a l of Educa­
tional Research. 19tl83-187, March, 1G? p .
Wall nee, W. L. "Fredictive Value of the ACE Psychological Examination,"
School ar.d Society, 70:?3’“^>5» July 9. 19^-c »
‘•Varies, EouglaB.
Proc edures in Hlgh School Teaching.
Macmillan Company» 19?7• 3^u or .

New York:

The

Yatson, I. "Significance of Hi^h School Marks for Predicting College
Success, " High S cl.n o 1 Tencher, 10
November, 1G~U.
Weinland, J. 0. "How Successful College Students Study," J o u m a l of
Educational Psycholog y , ^l:c.ri—
October, 1930.
Wetcel, William A. Biography of a High School .
Book Connany, 1°37»
7 on.

New York*

American

Whiurle, George M. "The Improvement of Educational Research," School
and Society. P 6 i P ^ - P ^ O . August P J , 1 9 P J •
Whitney, F. P.
"The ^:es tior.r.ai re Craze," Education Review, 6?:139“ i ^ »
October, 19p^.
Wiley, L. E. , and K. J. Sheridan, "Study of College Grades," Amerl can
As soclation of College Regi strars 1 J oum a l, 17 :?F— 3^, October, 19^1.
Williamson, E. G.
" Rel a t.i on s :i ^ of N u .fcer of Hours of Study to Sc h o l a r —
s h i o ," J o u r n a l of Ed .
’c^ t ional Psychology, P '~:p FP—
, December,

1935.
Wilson, H. h , ar.o G. M. W: Ison.
71he .lotiyation of Sc~
v o b : Houghton Mifflin ro n r a n y , 19'"!. ~ "'73 PP.

r

ol_ Wo r'<

new

1?7

Wilson, H. B. , G. C. Kyte, and H. G. Lull. Modern Methods of Teaching.
New York:
Silver, Burdette and Company" 19^4.
? % £ pp.
Wittenbom, J. R. , R. P. Larson, and R. L. Mogil.
"Empirical Evaluation
of Study Habits for College Courses in French and Soarish," Journal
of Ed ucatlonal Psychology,
i LLq— U7L1
., November, 19*43.
Woodring, M. N . , and C. W. Flemming.
"Directing Study Through the As—
signment," Teachers College Reco r d ,
^95* May, 193?.
Wrenn, C. G . , and W. J. Humber.
"Study Habits Associated with High and
Low Scholarship," J o u m al of Ed uca tional Psychology, 7-;3ll— £lS,
November, 19*41.
Wrinkle, William L . , and Winfield D. Armentroat.
Pirected Observation
end Teaching in Seconder:/ Schools.
New York:
The Macmillan Counanv,
1?7°. * 39° on.
Wylie, A. T.
"To What Extent May We Rely Upon the Answers to a School
Questionnaire?" Journal of Educational Method, 6 :
,
Feb n a r y ,
1927.
' ----------------Wynne, John P.
The Teacher and the Cnrriculum.
Hall, Inc., 1937h
EUO np.
Yoakum, G. A. Th.e Imnrovemen t of tj]_e As sl y m en t.
millan Coraoany, 191°.
Phh pi.

New York:
N e w York:

Prentice—
Trie Mac­